Thesis for the degree of Doctor of Philosophy
FLUORESCENT NUCLEOBASE ANALOGUES
AND THEIR USE FOR INVESTIGATING DNA
INTERACTIONS
ANKE DIERCKX
Department of Chemical and Biological Engineering
Chalmers University of Technology
Gothenburg, Sweden 2014
Fluorescent Nucleobase Analogues and their use for Investigating DNA
Interactions
ANKE DIERCKX
ISBN 978-91-7597-020-2
© ANKE DIERCKX, 2014
Doktorsavhandlingar vid Chalmers tekniska högskola
Ny serie nr 3701
ISSN 0346-718X
Department of Chemical and Biological Engineering
Chalmers University of Technology
SE-412 96 Gothenburg
Sweden
Telephone + 46 (0)31-772 1000
Cover: Left: Chemical structure of adenine analogues AT, qA and a member of the
A7T-family. Also shown is a putative structure of qA inside a B-DNA decamer.
Right: Tricyclic cytosine analogue FRET-pair tCO (purple) and tCnitro (orange) and
their framework placed into a B-DNA decamer with a separation of 6 bp.
Chalmers Reproservice
Gothenburg, Sweden 2014
ii
FLUORESCENT NUCLEOBASE ANALOGUES AND THEIR
USE FOR INVESTIGATING DNA INTERACTIONS
Anke Dierckx
Department of Chemical and Biological Engineering
Chalmers University of Technology
Abstract
Ever since unravelling the structure of DNA, an expanding research field has
emerged with ongoing efforts dedicated to increase our understanding of the
molecule of life. Since the natural nucleobases are virtually non-emissive, it has
been a challenge for decades to ‘light up’ DNA/RNA in order to investigate their
properties utilizing fluorescence techniques. This thesis focuses on fluorescent
nucleobase analogues (FBAs) as probes for fluorescently labeling DNA and
investigating its interactions, for example, with proteins. These artificial
nucleobases attempt to closely mimic the characteristics of natural bases, while
introducing fluorescence properties to the system.
The first part of this work comprises the characterization of both new and
established FBAs. Photophysical and base-mimicking properties of two
fluorescent adenine analogues, triazole adenine (AT) and quadracyclic adenine
(qA) are presented. Both exhibit promising features compared to the widely used
commercially available adenine mimic 2-aminopurine (2-AP). Even though AT
shows promising emission as a monomer and in certain DNA surroundings, it
destabilizes the B-DNA duplex structure, most likely due to its C8-triazole
extension. In order to overcome this effect, a new family of triazole adenine
analogues extended on the 7-position was synthesized and photophysically
characterized. The second thoroughly characterized adenine analogue, qA, is
moderately fluorescent both as a monomer and inside DNA but in contrast to AT,
the two-ring extension on qA is suggested to be well accommodated in the major
groove and renders the DNA-duplex unperturbed or even stabilized, depending on
the surrounding sequence. Finally, the photostability of tC, an already established
FBA of the tricyclic cytosine family, was investigated. The latter yields a single
photoproduct with a decreased fluorescence, which destabilizes DNA duplexes.
In the second part of this work, the tricyclic cytosine FBA FRET-pair, tCOtCnitro was applied in exploring the role of the mammalian transcription factor A in
mitochondrial transcription. Furthermore, it was called upon to help resolve the
order of events in which the different components of the transcription machinery
initiate transcription.
Keywords: Fluorescence quantum yield, B-DNA, fluorescent nucleobase
analogue, duplex stability, circular dichroism, adenine analogue, FRET
iii
List of Publications
The thesis is based on the work described in the following articles, referred to by
Roman numerals in the text:
I
Anke Dierckx, Peter Dinér, Afaf H. El-Sagheer, Joshi Dhruval Kumar,
Tom Brown, Morten Grøtli and L. Marcus Wilhelmsson
Characterization of photophysical and base-mimicking properties of a
novel fluorescent adenine analogue in DNA
Nucleic Acids Research, 2011, 39, 4513-4524.
II
Christopher P. Lawson§, Anke Dierckx§, Francois-Alexandre Miannay,
Eric Wellner, L. Marcus Wilhelmsson and Morten Grøtli
Synthesis and photophysical characterization of new fluorescent triazole
adenine analogues
Submitted to Organic & Biomolecular Chemistry
§
III
Both authors contributed equally to this work
Anke Dierckx§, Francois-Alexandre Miannay§, Nouha Ben Gaied, Søren
Preus, Markus Björck, Tom Brown and L. Marcus Wilhelmsson
Quadracyclic adenine: A non-perturbing fluorescent adenine analogue
Chemistry A European Journal, 2012, 18, 5987-5997.
§
Both authors contributed equally to this work
IV
Søren Preus, Søren Jønck, Michael Pittelkow, Anke Dierckx, Thitinun
Karpkird, Bo Albinsson and L. Marcus Wilhelmsson
The photoinduced transformations of fluorescent DNA base analogue tC
triggers DNA melting
Photochemical & Photobiological Sciences, 2013, 12, 1416-1422.
V
Yonghong Shi, Anke Dierckx, Paulina H. Wanrooij, Sjoerd Wanrooij,
Nils-Göran Larsson, L. Marcus Wilhelmsson, Maria Falkenberg and
Claes M. Gustafsson
Mammalian transcription factor A is a core component of the
mitochondrial transcription machinery
Proceedings of the National Academy of Sciences, 2012, 109, 1651016515.
VI
Viktor Posse, Emily Hoberg, Anke Dierckx, Saba Shahzad, Camilla
Koolmeister, Nils-Göran Larsson, L. Marcus Wilhelmsson, B. Martin
Hällberg and Claes M. Gustafsson
The amino terminal extension of mammalian mitochondrial RNA
polymerase ensures promoter specific transcription initiation
Nucleic Acids Research, 2014, 42, 3638-3647.
iv
Contribution Report
Paper I:
Main
responsible
for
performing
characterization and writing the paper.
Paper II:
Main
responsible
for
performing
the
photophysical
characterization. Wrote the paper together with C.P.L..
Paper III:
Performed photophysical characterization and wrote paper
together with F.A.M.. TDDFT calculations were done by S.P. and
quantum yield measurements on the qA monomer by F.A.M..
Paper IV:
Performed fast photoconversion of tC in duplex DNA, recorded
melting curves and CD spectra. Proofread paper.
Paper V:
Performed and analyzed spectroscopic measurements. Involved in
writing corresponding paragraphs and proofread paper.
Paper VI:
Performed and analyzed spectroscopic measurements. Proofread
paper.
v
the
photophysical
vi
Table of Contents
1
Introduction ..........................................................................................1
2
Background ..........................................................................................5
2.1
Deoxyribonucleic acid .......................................................................5
2.2
Fluorescence and DNA ......................................................................8
2.2.1
DNA ligands .............................................................................9
2.2.2
Covalent modifications .............................................................9
3
Theory and Methodology ..................................................................13
3.1
Interaction of light and matter .........................................................13
3.2
Absorption .......................................................................................15
3.3
Circular dichroism ...........................................................................16
3.4
Fluorescence ....................................................................................17
3.4.1
Fluorescence principles ...........................................................17
3.4.2
Steady state fluorescence ........................................................20
3.4.3
Steady state fluorescence anisotropy.......................................21
3.4.4
Time-resolved fluorescence ....................................................22
3.4.5
Förster resonance energy transfer (FRET) ..............................24
3.4.6
The tricyclic cytosine FRET-pair ............................................26
3.5
4
DNA melting ...................................................................................28
Results .................................................................................................31
4.1
Characterization of fluorescent nucleobase analogues ....................31
4.1.1
Characterization of fluorescent adenine analogues .................31
4.1.2
Photostability of the tricyclic cytosine analogue tC ................44
4.2 Application of the tricyclic cytosine FRET-pair to investigate
protein-DNA interactions ..................................................................................46
5
Conclusion and outlook .....................................................................51
6
Acknowledgements.............................................................................55
7
References ...........................................................................................57
vii
viii
1 Introduction
In 1869 a young Swiss physician, Friedrich Miescher, discovered a
phosphorous-containing substance in the nuclei of leucocytes, which he named
nuclein. The latter mainly consists of chromatin, a complex of chromosomal
proteins and deoxyribonucleic acid, DNA, later identified as the carrier of our
genetic information.[1,2] A major breakthrough towards better understanding its
structure and therefore its way of function came in 1953 when Watson and Crick
proposed the famous double helical structure, in which two complementary
strands consisting of the four natural nucleobases (adenine, guanine, cytosine and
thymine), deoxyribose sugar and phosphate groups are paired in the shape of a
right winding corkscrew.[3] Another great development in nucleic acids research
was the automation of DNA synthesis, allowing short (<200 bp) synthetic
oligonucleotide sequences to be ordered from various commercial sources
nowadays at low costs.[4,5] Furthermore, the prestigious ‘human genome project’,
which aimed at extracting the sequence of the entire human genome, was enabled
due to the development of DNA sequencing methods.[6] Even though much
progress has been made, there are many aspects yet to be understood about this
surprisingly simple, yet so important molecule. Ribonucleic acid (RNA) has also
attracted much scientific interest as it has become apparent that it is not merely a
messenger between our genetic information and proteins. In fact, important roles
have been suggested for non-coding RNA sequences in gene regulation and other
regulatory processes in the cell.[7-10]
A quick search on the world wide web revealed that DNA is referred to
nowadays as the instruction manual of every living cell or the blueprint of life to
name but a few, thus it is no surprise that this molecule still attracts massive
research attention today. Furthermore it has also generated substantial interest as a
building block for nano-constructs due to its well-ordered structure and
predictable assembly.[11-13]
Since many complex processes are required for both DNA and RNA to
execute their biological functions, it is a field of ongoing interest to provide a
better understanding of the structure and dynamics of these molecules. Another
major area of interest is the provision of clear insights into the complex network of
interactions of DNA with other molecules such as proteins or RNA.
Besides a quest for better understanding the processes DNA and RNA are
involved in, there is also a broad interest in improving the techniques which allow
the retrieval of such information. There are different approaches to study the
above mentioned aspects of DNA and RNA, such as the high resolution methods
x-ray crystallography[14] and Nuclear Magnetic Resonance (NMR)[15]. Although
1
powerful at providing detailed structural information, a limitation of x-ray
crystallography lies in the need for obtaining a crystallized sample, which hardly
ever is a trivial task. Furthermore, results obtained for a crystalline molecule may
differ significantly from the physiological conditions in which it is usually found.
NMR, on the other hand, can be recorded in solution and is therefore a powerful
tool to derive solution structures of nucleic acids. Additionally, it can yield
dynamic information about the molecule. However, NMR suffers from the
increasing complexity of the obtained signal with molecule size.
Another powerful tool for retrieving information about DNA and RNA is
fluorescence, which is very sensitive, straightforward and requires only small
amounts of sample in contrast to the above named methods. Moreover, it allows
for real-time measurements in solution. Numerous proteins have been studied
taking advantage of the fluorescence of the amino acid tryptophan.[16] The
nucleobases, the natural building blocks of DNA and RNA, are however virtually
non-fluorescent (Φf~10-4, τf <1 ps in aqueous solution at room temperature)[17,18],
which is probably a convenient way for nature to protect us from genetic damage
caused by UV-light.[19] As a result, DNA needs to be labelled with a fluorophore
in order to study its properties using fluorescence techniques, thereby allowing a
good signal to noise ratio due to negligible background contributions from the
natural bases.
There are different strategies for fluorescently tagging nucleic acids, each
with specific advantages depending on the aim of the study. Easy non-covalent
labelling can be achieved by groove binders and intercalators.[20-23] Covalent
modifications are often achieved by tethering bright commercially available dyes
(for example Cy-dyes, fluorescein and Alexa-dyes) to the DNA via a linker,
allowing studies down to the single-molecule level.[24-26]. However, a number of
intercalating dyes usually modify and elongate the DNA structure since they need
to fit into the base-stack. Also bulky external dyes can interfere with for example
DNA-protein interactions due to steric hindrance. Furthermore, the emission of
external chromophores is usually insensitive to subtle structural changes inside the
DNA. This is where fluorescent nucleobase analogues (FBAs) come in handy.
These probes are significantly fluorescent (even though less bright than
commercially available external dyes) and mimic the natural nucleobases, thereby
minimally perturbing the duplex structure. FBAs can be specifically incorporated
into DNA and for most members, their emission is sensitive to changes in the
direct environment so they can report on local structural alterations in the DNA
duplex.[27-31]
One of the most widely utilized FBAs to date, 2-aminopurine (2-AP), was
among the first to be discovered (Figure 1).[32] 2-AP is an adenine isomorph that is
2
highly fluorescent as a monomer in solution, but is significantly quenched inside
single- and double-stranded nucleic acid systems.[32] It moderately destabilizes the
DNA duplex[33] and can form base-pairs with not just thymine but also cytosine,
albeit less stable.[34-36] Different classes of base analogues have been developed
and explored over the last few decades, each with its own specific qualities.
However, their design remains problematic due to the difficulties associated with
prediction of emissive behaviour based on chemical structure. The development of
FBAs with improved properties, e.g. higher fluorescence quantum yields, higher
extinction coefficients, selective base-pairing and which cause minimal or no
destabilization is an area of ongoing interest. Therefore the development and
thorough characterization of new FBAs is highly significant to the field.
The focus of the first part of this thesis is concerned with the spectroscopic
characterization of both new and already introduced FBAs (papers I-IV). Two
fluorescent adenine analogues, triazole adenine (AT) and quadracyclic adenine
(qA) were thoroughly studied both as monomers and inside DNA (Figure 1). Due
to its good emissive properties reported in a previous study,[37] AT was
investigated as a nucleoside and incorporated into oligonucleotides with different
base surroundings (paper I). AT shows promising emissive properties evidenced
by high quantum yields it displays in DNA depending on the base-surroundings.
However, it causes a moderate destabilization of the duplex, which can probably
be attributed to the C8 modification. The latter could force AT to adopt a syn
conformation around the glycosidic bond. As a follow-up study, trying to
overcome this undesired destabilizing effect, a new series of triazole adenine
analogues (A7T-family) was designed, bearing the modification on the purine 7position instead (paper II, Figure 1).
In an attempt to discover an adenine equivalent to the promising previously
characterized tricyclic cytosines tCO, tC and tCnitro, [38-42], an extended quadracyclic
adenine analogue, qA, was investigated (paper III, Figure 1). Even though qA
does not show the stable emission inside DNA as was reported for both tC[38] and
tCO[39], it demonstrates a similar increase or preservation of duplex stability
depending on the nature of the surrounding bases.
Finally, to approach the idea of FBAs for single-molecule studies, a good
understanding of their photostability is vital. Therefore, the photodegradation of
the tricyclic cytosine analogue tC (Figure 1) was characterized as a monomer and
inside DNA. This lead to the surprising discovery that intense irradiation can
trigger its photoconversion to a product which destabilizes DNA duplexes
(paper IV).
The second part of this work is concerned with the applications of FBAs,
more specifically the FBA FRET-pair, tCO-tCnitro (Figure 1), in investigating
3
mammalian mitochondrial transcription (papers V-VI). A study by Börjesson et al.
published in 2009 reported on the responsiveness of this FRET-pair to alterations
in both distance and orientation due to its rigid positioning in the DNA duplex.[40]
The system is used here to report on local structural alterations in the DNA close
to the transcription start site upon binding of the mammalian mitochondrial
transcription factor A (TFAM) (paper V). In a follow-up study, the FRET-system
was also used to help determine the sequence of events caused by TFAM, the
TFB2M-factor and human mitochondrial RNA polymerase (POLRMT) at the
promoter to initiate transcription (paper VI).
Figure 1 Structure of 2-aminopurine (2-AP) and the adenine analogues discussed
in this work: AT (paper I), the A7T-family (paper II) and qA (paper III). Also the
already established tricyclic cytosine family is shown (studied/applied in papers
IV-VI). The backbone of natural adenine/cytosine is shown in grey.
4
2 Background
Since DNA is the central molecule in this thesis, its structure and function
will be discussed. Fluorescent probes, more specifically fluorescent base
analogues, will also be highlighted in this chapter as tools to study different
aspects of DNA and RNA.
2.1 Deoxyribonucleic acid
In the 20th century, molecular genetics greatly advanced our understanding of
the composition and structure of genes as well as how they exert their function. As
mentioned in the introduction, it was the young Swiss physician Friedrich
Miescher who discovered the phosphorus-containing substance nuclein in the
nucleus of white blood cells in 1869.[2] However, it was not until 1944 that
Oswald Avery, Colin MacLeod and Maclyn McCarty reported a transformation
experiment that provided the proof that DNA, one of the main components of
nuclein, is the material that carries genetic information.[43] By then it was also
known that DNA was built of nitrogenous bases (adenine (A), cytosine (C),
guanine (G) and thymine (T)), as well as phosphoric acid and the sugar
deoxyribose (Figure 2). Adenine and guanine are referred to as purines whereas
cytosine and thymine are referred to as pyrimidines according to the parent
molecules they are related to. One main piece of the puzzle left to discover by the
end of the 1940s was the structure of DNA and how it executes its function as
gene carrier.[1]
In 1950 Erwin Chargaff found the amount of purines and pyrimidines in DNA
from various sources to be roughly equal.[44] Furthermore, this also was true for
the amounts of adenine and thymine as well as guanine and cytosine. Another
crucial piece of information came from x-ray fibre diffraction patterns made by
Rosalind Franklin and Maurice Wilkins.[45,46] This lead James Watson and Francis
Crick to link the information together and publish their famous paper ‘Molecular
structure of nucleic acids’ in 1953, in which they describe the double helical
character of DNA.[1,3]
The DNA double helix is built up of nucleotides, consisting of a nucleoside (a
nucleobase linked to β-D-2’-deoxyribose via a glycosidic bond) and a phosphate
group. These nucleotides are linked together by phosphodiester bonds to form a
polynucleotide or single-stranded DNA chain. Usually, the nucleotides adopt the
anti-conformation about the glycosidic bond, which means that the base is turned
5
away from the sugar ring to avoid steric hindrance between the sugar and either
N3 in purines or the carbonyl oxygen in pyrimidines (Figure 2a). A single DNA
sequence is able to hybridize to a complementary strand, having the sequence of
bases running in the opposite direction (antiparallel), creating a double helix. This
means that at both ends of the duplex, the 5’-phosphate group of the last
nucleotide of one strand and the 3’-hydroxyl group of the other strand can be
found. The complementary strand contains a sequence of nucleotides so that the
nucleobases adenine and thymine as well as cytosine and guanine are always
paired between both strands. This means that each sequence contains the
information necessary to synthesize a complementary strand.[1,47,48]
Figure 2 (a) Dinucleotide of 2’-deoxyadenosine linked to 2’-deoxycytidine by a
phosphodiester bond. (b) Base pairing pattern between the natural nucleobases
guanine (G) and cytosine (C) as well as between thymine (T) and adenine (A). Also
the major-and minor groove sides of each base-pair are indicated. R indicates 2’deoxyribose.
Hydrogen bonds are formed between the complementary bases contained in
both strands of the DNA duplex: two hydrogen bonds between each adeninethymine pair and three between cytosine and guanine (Figure 2b). Furthermore, ππ-stacking (a form of van der Waals interaction) and hydrophobic interactions
between the aromatic nucleobases contribute significantly to the duplex stability.
At physiological pH, the DNA molecule is a large polyanion because of the
negatively charged phosphate groups in its backbone. This is why salt
concentration has a large impact on duplex stability, since positively charged salt
ions such as Na+ or Mg2+ can shield the negative backbone charges, decreasing the
6
repulsion between both strands. Also duplex length, sequence, type of solvent and
pH play a vital role in the stability of a DNA molecule. The DNA duplex can
adopt different conformations which are greatly influenced by hydration and ions.
In general DNA double helices are found in the B-form under physiological
conditions in aqueous solution (Figure 3), characterized by a right winding helical
twist about 36° and a 3.4 Å rise between the base-pairs. About 10-10.6 base-pairs
are contained per helical turn (depending on local sequence).[1,47,48] Along the
surface of the helix two distinct grooves can be detected, which are similar in
depth but different in width, known as the major (11 Å wide) and minor (6 Å
wide) groove.[49] Other helical forms that exist are Z-DNA and A-DNA. The latter
also is a right-winding helix, but more compressed than B-DNA, which can be
obtained in dehydrated samples and DNA-RNA hybrids.[47,48] Z-DNA on the other
hand is a left winding helix and can be formed mainly in alternating purinepyrimidine sequences such as poly[dG-dC]·poly[dG-dC] at high ionic
strength.[47,48,50] Several lines of evidence, such as the identification of Z-DNA
binding proteins, suggest a biological role for this helical form in vivo.[50] Also
other non-B-DNA forms have been identified, such as the G-quadruplex, triplex
and i-motif to name a few.[50]
Figure 3 B-DNA double helix. Schematic representation (left) as well as spacefilling model (right) with minor and major groove indicated.
After describing its basic structure, the function of DNA and connection to
other processes in the cell will be briefly discussed here. First of all it is important
to mention that the DNA molecule occurs in different forms depending on cell
7
type. Prokaryotes have a single, typically circular DNA molecule, which is
associated with proteins. Eukaryotes, on the other hand, have their DNA organized
in chromosomes, which are highly ordered complexes of the DNA double helix
with proteins, such as histones. This means that complex dynamic processes must
be involved for the information contained in the DNA sequence to become
available. The central dogma of information transfer in cells can be summarized as
follows: the sequence of a gene coded in DNA is transcribed into RNA, for
example messenger RNA (mRNA) which is then processed and translated by
ribosomes into proteins which can execute a specific function in the cell (Figure
4).[1] The process of transcription in mammalian mitochondria is investigated in
paper V and paper VI. Of course this is not the only process in which the genetic
code needs to be read. Also upon cell division, all the genetic information needs to
be replicated in order to provide both daughter cells with the same instruction
manual. All these complex processes require interactions of DNA with a whole
army of proteins such as transcription factors, polymerases and other molecules. It
is apparent that dynamic DNA processes are constantly occurring in cells and that
our genetic material largely exceeds the role of a passive information carrier.
Figure 4 Schematic representation of replication and transcription of DNA as well
as translation of messenger RNA.
2.2 Fluorescence and DNA
Due to the virtually non-emissive nature of the natural DNA/RNA building
blocks, [17,18] fluorophores need to be introduced into these systems in order to
study their properties using fluorescence. In the introduction, main strategies to
8
render DNA fluorescent were briefly touched upon. This section divides these into
non-covalent (DNA ligands) and covalent modifications of which fluorescent
nucleobase analogues (FBAs) receive special attention as covalent modifications
according to their relevance for this thesis.
2.2.1 DNA ligands
DNA can be easily labelled non-covalently with fluorophores that bind more
or less non-specifically by hydrophobic and/or by electrostatic interactions with
the negatively charged phosphate backbone. DNA ligands are usually planar,
positively charged aromatic molecules (Figure 5). Furthermore, their quantum
yield is typically increased several orders of magnitudes upon binding to DNA,
yielding a good signal to noise ratio. Intercalators, such as oxazole yellow (YO),
its homodimer YOYO[23] and ethidium bromide [20] insert themselves between two
DNA bases, thereby elongating the duplex to some extent (Figure 5). Other
examples of DNA ligands are groove binders such as 4’,6-diamidino-2phenylindole (DAPI)[21] and Hoechst[22]. Often, DNA ligands are used as an easy
stain for visualizing DNA after electrophoresis or in microscopy. They are also
used in artificial light harvesting in DNA nanoconstructs.[51] However, the binding
is rather non-specific and is a distribution, which makes it difficult to know where
and exactly how many chromophores that are attached to each DNA duplex.
Figure 5 Examples of DNA ligands such as the intercalators ethidium and YO as
well as the groove binder DAPI.
2.2.2 Covalent modifications
The most common way of covalently labelling nucleic acids is by tethering an
external probe to it with a flexible hydrocarbon linker. In this way almost any
commercially available bright dye can be attached as a step during solid-phase
DNA synthesis or as a post-synthetic step. Examples are fluorescein,
carboxytetramethylrhodamine (TAMRA) and the series of cyanine dyes or Alexa
9
dyes (Figure 6). Their brightness allows detection down to single-molecule
level.[24-26] However, as for a number of intercalating dyes, bulky external
fluorophores can perturb the DNA and its interactions with other molecules, such
as proteins or other ligands. Dyes have been shown to interact with DNA, for
example through electrostatic interactions with the backbone, intercalation or endstacking.[52-56] Furthermore, the long linkers frequently used can contribute to dye
diffusion and reorientation,[53,56,57] increasing the difficulty of interpreting
anisotropy or quantitative FRET experiments. Additionally, the positioning of the
dye outside of the base-stack limits the retrieval of local site-specific information.
Figure 6 Examples of covalent external dyes for DNA (top row) as well as of
previously characterized FBAs (bottom two rows), for which the basic skeleton of
the natural nucleobase is shown in grey.
The main focus of this thesis is on the use of FBAs, covalently incorporated
base-mimics, for rendering DNA fluorescent. In this work, the term FBAs is used
for molecules that closely resemble one of the natural nucleobases in terms of
hydrogen bonding capacity, while minimally perturbing the duplex integrity
(conformation and stability) and at the same time introducing fluorescence into the
nucleic acid system. Even though FBAs cannot compete with the external
10
commercial fluorophores in terms of brightness, they can be incorporated close to
or at the very site of interest in the DNA helix, while preserving the native
conformation of the DNA. The emission of most FBAs is sensitive to changes in
their direct environment, making them responsive to local alterations.
Furthermore, their more rigid incorporation into the base-stack allows them to
report on the motion of the DNA helix, rather than the dye itself.
Over the last few decades, FBAs for each of the nucleobases have been
reported and to date this research field continues to evolve rapidly (for a detailed
overview see [27-31,58]). Depending on the type of experiment, different FBAs may
be more suitable, based on the sensitivity of their emission to duplex structure,
hydrogen bonding (with a complementary/mismatch base), environment (pH, base
stacking), interactions of the DNA with other molecules such as polymerases, etc.
However, the design of FBAs for each specific need has proven challenging. One
faces structural constraints in order to leave the natural B-DNA as unperturbed as
possible and preserving the hydrogen-bonding capacity of the base. Furthermore,
there is a poor understanding of the correlation between structure and emissive
properties of these molecules. Frequently, FBAs show a significant decrease in
fluorescence quantum yield upon incorporation into nucleic acid systems.
Exceptions are the adenine analogues 2-(3-phenylpropyl)adenosine (A-3CPh) and
2-(4-phenylbutyl)adenosine (A-4CPh), which show an increase in quantum yield
upon incorporation into RNA[59] as well as the tricyclic cytosines tC and tCO,
whose emission is relatively stable[38,39].
The pteridines are a well-studied group of FBAs with the guanine analogues
6-MI and 3-MI[60] as well as the adenine analogues 6MAP and DMAP[61] as the
most promising members (Figure 6).[27,62] Although being moderately fluorescent
inside DNA, they cause a destabilization of the duplex (with exception of 6-MI),
similar to a single base pair mismatch for 3-MI.[61,62] Other groups of FBAs
comprise the pyrimidine analogues developed by Tor et al.,[63-67] as well as their
recently developed emissive RNA alphabet (thA, thG, thC and thU)[68] (thA, thC
shown in Figure 6). The base-discriminating fluorescent (BDF) nucleosides
designed by Saito and co-workers could, as the name suggests, be applied for
analysing single nucleotide polymorphisms (SNP).[58] One example is the cytosine
analogue benzopyridopyrimidine (BPP), which can form stable base-pairs with
both adenine and guanine (Figure 6). Incorporation of a vinyl group on C8 of
adenine produces an environmental sensitive fluorescent nucleobase, 8-vinyldeoxyadenosine (8vdA), which is only minimally perturbing the duplex structure
(Figure 6).[69] The size-expanded DNA alphabet designed by Kool et al., although
perturbing the DNA duplex, has been practical for investigating steric effects in
DNA.[70-72] Pyrrolo-dC, a bicyclic cytosine analogue, hybridizes selectively with
guanine and preserves duplex stability (Figure 6). It was formed by accident
11
during solid-phase synthesis upon attempting to incorporate the analogue furanodT into DNA.[73] A number of pyrrolo-dC-derivatives have also been
developed.[74-77]
One of the first successful and still most widely used FBAs is the adenine
analogue 2-aminopurine (2-AP, Figure 6)[32], which over the years has found its
way into countless applications involving DNA and more recently also RNA.[28-30]
Its lowest energy absorption band is redshifted compared to natural adenine (~ 303
nm in water), allowing for selective excitation, whereas its emission is centred
around 370 nm. Even though 2-AP shows a high fluorescence quantum yield in
solution (Φf=68% in water), it is quenched almost a hundred fold upon
incorporation into nucleic acids, depending on the sequence environment.[32,78] As
was mentioned above, this is a common property for FBAs. Furthermore 2aminopurine forms base-pairs with not only thymine, but also with cytosine, albeit
less stably.[34-36] As has been reported for other FBAs, 2-AP moderately
destabilizes the DNA duplex [33].
Over the past decades FBAs have been used in numerous applications, of
which only a few will be highlighted here (for a more detailed overview see [27-31]).
Most of these exploit the sensitivity of the FBAs’ emission to their
microenvironment. Applications involve the investigation of certain aspects of
DNA structure or dynamics. 6MAP has for example been used to analyse the
premelting transitions of DNA A-tracts.[79] Other examples involve 2-AP in
studies concerning the effect of cations and sequence on base-stacking interactions
at abasic sites [80] or the dynamics of 2-AP mismatches in DNA[81]. A second type
of applications puts more focus on the interaction of DNA with other molecules.
This is illustrated nicely by the incorporation of 2-AP in the promoter sequence to
which bacteriophage T7 RNA polymerase binds, yielding information concerning
kinetics of the promoter binding and open complex formation.[82] Another
application is the use of the guanine analogue 3-MI to study HIV-1 integrase.[83,84]
Recently, FBAs also started making their way into the field of nanotechnology. A
first report was published in 2009 and involves the fluorescent cytosine analogue
tCO, reporting on the local stability of self-assembling DNA hexagons.[85] FBAs
are also becoming more and more important in the expanding field of RNA. An
example is the use of 2-AP to study riboswitches, non-coding RNA elements that
control gene expression, as a result of binding of small molecules.[86,87] 6phenylpyrrolocytosine (PhpC) was applied to follow cellular trafficking of
siRNAs, whose gene silencing activity was virtually unaltered compared to natural
siRNA.[88] Also protein-RNA interactions have been studied with the help of
FBAs. A nice illustration is a study concerning the change in emissive properties
of 2-AP during the RNA editing process carried out by adenosine deaminases
acting on RNA (ADARs).[89,90]
12
3 Theory and Methodology
In this chapter the methods used to perform the experiments presented in this
thesis will be introduced as well as their underlying theory. To begin with, the
interaction of light and matter will be discussed briefly, leading to the description
of absorption, circular dichroism and fluorescence spectroscopy. Furthermore, a
section is dedicated to the principle of Förster Resonance Energy Transfer (FRET)
and its application in DNA. Experimental details are described in the according
sections of the published papers included in this thesis. For a thorough description
of photochemistry[91-93], circular dichroism[94] and fluorescence techniques [16,93,95],
the reader is referred to the corresponding references.
3.1 Interaction of light and matter
The interaction of electromagnetic radiation with matter provides
spectroscopists with desired information about the molecules or atoms they study.
In order to describe this interaction, electromagnetic radiation must be treated as
having a wave as well as particle character. In terms of classical physics, the
electromagnetic field can be seen as a harmonic wave of an oscillating electric and
magnetic field, travelling at the speed of light (c=3.108 m s-1). The electric and
magnetic fields oscillate in phase with a wavelength λ and frequency v and
describe waves which are perpendicular to each other and to the direction of
propagation. The radiation can, however, also be described as a flow of photons,
which can be seen as small energy packages with energy
‫ ܧ‬ൌ ݄‫ݒ‬
(1)
where h is Planck´s constant (h=6.626.10-34 J s) and v is related to the speed of
light, c, as
‫ݒ‬ൌ
௖
(2)
ఒ
Molecules usually exist in discrete energy levels (i.e. rotational, vibrational
and electronic energy levels). This means that a particular energy gap has to be
overcome to enable a promotion to a higher energy level. A molecule in state m
can get excited to a state n if it absorbs a photon which holds exactly the amount
of energy equal to the energy difference between both states. This is known as
Bohr´s frequency condition and is given as
13
ο‫ ܧ‬ൌ ‫ܧ‬௡ െ ‫ܧ‬௠ ൌ ݄‫ݒ‬
(3)
Quantum mechanics provides a theory to describe the energy levels that a
molecule can occupy and to predict which transitions can occur between these
levels. The state of a system (e.g. a molecule) is described by a wavefunction ߖ.
The oscillating electric field of radiation (of the correct frequency) can cause an
oscillation in the electron cloud of a molecule, initially in state ߖ௠ and thereby
exciting it to a higher electronic state described by ߖ௡ . The transition will
preferentially occur if the electric field of the radiation is polarized parallel to the
polarization caused in the electron cloud during the process. This dipole induced
by light determines the magnitude of the probability of this transition occurring
ሬԦ௡௠ , which is given by:
and is called the transition dipole moment, Ɋ
Ɋ
ሬԦ௡௠ ൌ ‫ߖ ׬‬௡‫ כ‬Ɋොߖ௠ ݀߬
(4)
where the integral is taken over all space (dτ=dxdydz) of the product of the
complex conjugate of the wave function of state n (ߖ௡‫ ) כ‬with the electric dipole
moment operator, Ɋො, and the wave function of the initial state ߖ௠ . The electric
dipole moment operator is described as
Ɋො ൌ σ௜ ‫ݍ‬௜ ‫ݎ‬Ԧ௜
(5)
for which qi and ‫ݎ‬Ԧ௜ are the charge and position vector of the ith electron
respectively.
The probability for the transition occurring from state m to n is proportional to
the square of the magnitude of the transition dipole moment, also known as the
dipole strength, Dnm
ܲ௡௠ ‫ܦ ן‬௡௠
(6)
where
‫ܦ‬௡௠ ൌ ȁߤԦ௡௠ ȁଶ
(7)
The dipole strength is also related to an experimentally determinable
parameter, the absorption coefficient ߝሺ‫ݒ‬ሻ (M-1cm-1) (also called extinction
coefficient)
‫ ߝ ׬‬ሺ‫ݒ‬ሻ݀‫ݒ ן ݒ‬௡௠ ‫ܦ‬௡௠
(8)
14
where ‫ݒ‬௡௠ is the transition frequency and ߝሺ‫ݒ‬ሻ gives the frequency dependence of
the absorption band.
Another measure of the intensity of the transition from states m to n at a
frequency ‫ݒ‬௡௠ is the oscillator strength fnm, which represents the ratio of the
intensity of the absorption to that expected from a three dimensional harmonic
oscillator. The oscillator strength can be related to the dipole strength as well as to
the integrated absorption coefficient:
݂௡௠ ‫ݒ ן‬௡௠ ‫ܦ‬௡௠ ‫ ߝ ׬ ן‬ሺ‫ݒ‬ሻ݀‫ݒ‬
(9)
3.2 Absorption
The absorption of a sample is a measure for the amount of photons it absorbs
upon exposing it to radiation of a suitable wavelength. Usually absorption is
measured in a spectrophotometer, for which the basic components are a light
source, a monochromator and a detector (Figure 7).
Figure 7 Basic setup of a spectrophotometer. Radiation of the correct wavelength
is selected from the light source by a monochromator and led through air (I0) or
through a sample with length ݈ (I) onto a detector.
The sample is placed in the light emitted by a source and its absorbance at a
certain wavelength is related to the ratio of the intensities of the incoming (I0) and
transmitted radiation (I). It is also related to the absorption coefficient ߝሺߣሻ, the
concentration (c) and the path length (l) of the sample (often in a quartz cuvette).
This relation can be described as
బ ሺഊሻቁ ൌ ߝሺߣሻ݈ܿ
‫ܣ‬ሺߣሻ ൌ Ž‘‰ ቀ಺಺ሺഊሻ
( 10 )
15
Absorption spectroscopy is often performed to determine the concentration of
a sample with a known extinction coefficient ߝሺߣሻ but also allows the retrieval of
structural or conformational information. For new molecules, the absorption
coefficient can be determined and suitable excitation energies for fluorescence
measurements can be identified by making a simple absorption measurement.
3.3 Circular dichroism
The peculiar property of chiral molecules (i.e. molecules without a reflection
plane) to display an unequal absorption of left and right circularly polarized light
is known as circular dichroism (CD). A CD signal can also be induced for achiral
chromophores upon incorporation into a chiral environment, such as for some
DNA intercalators or groove binders. CD can be calculated as
‫ܦܥ‬ሺߣሻ ൌ ‫ܣ‬௟ ሺߣሻെ‫ܣ‬௥ ሺߣሻ ൌ ሺߝ௟ ሺߣሻ െ ߝ௥ ሺߣሻሻ݈ܿ ൌ οߝሺߣሻ݈ܿ
( 11 )
where οߝሺߣሻ is the molar circular dichroism (M-1cm-1) and c and l are the
concentration and path length of the sample respectively.
For biomolecules in solution, CD is often used to identify their secondary
structure.[94] The alignment of chromophores in a certain conformation of the
molecule may cause shifted or split transitions due to exciton interactions,
resulting in a specific CD signature. The CD spectrum of proteins, for example,
varies depending on their composition of β-sheets, α-helices and random coils.
Similarly, a DNA molecule in the A, B or Z conformation can be distinguished
based on characteristic CD features. The UV-absorption band for DNA is mainly
initiated by π-π* transitions of the nucleobases. However, the bases themselves are
achiral and thus do not possess a CD signal. The CD spectrum detected for DNA
is due to the linkage of the bases with chiral sugars and to their helical stacking.
Between 190 and 300 nm, the DNA CD arises mainly from the relative orientation
of the bases to each other. For the most common helical form, B-DNA, this results
in a CD band which is positive at ~275 nm, zero at ~258 nm and continuing to
negative at ~240 nm. Towards the higher energy region, the spectrum continues
with a band that increases towards less negative or positive at 220 nm, followed
by a small negative peak and finally a large positive peak between 180-190 nm.
A-DNA and double-stranded RNA display a similar spectrum as B-DNA,
however, the positive peak at ~275 nm is shifted towards 260 nm and increased in
intensity. The negative peak between 230-250 nm is less intense and there is an
16
intense negative band at 210 nm followed by a very intense positive band at 190
nm. The left-winding Z-DNA duplex, which can, for example, be obtained for
poly[dG-dC]·poly[dG-dC] at high ionic strength has a negative CD signal at 290
nm, a positive band at 260 nm and a large negative signal around 195-200 nm,
while passing through zero between 180-185 nm. Representative spectra of B- and
A-form DNA are shown in Figure 8.
Figure 8 Representative CD spectra for a 1 μM solution of a 38-mer DNA duplex
in the B-form (grey, 0.33 mM Na+, 0.18 mM phosphate buffer) or A-form (black,
0.33 mM Na+, 0.18 mM phosphate buffer, 78 % ethanol). The unit M-1 on the
vertical axis refers to the duplex concentration. Spectra were smoothed by
adjacent averaging over 5 points. The full sequence reads
5’-d(CCATCCCACCACGAGAGAGAGAGACGTCACCACCCTCC)-3’.
3.4 Fluorescence
In the previous paragraphs absorption was discussed as a process that can
render a molecule in an electronically excited state. After that, the molecule has
several options for losing its excess energy and returning to the ground state. One
of these routes includes fluorescence, which will be discussed in the following
paragraphs along with methods to study these properties and their applications.
3.4.1 Fluorescence principles
Fluorescence is the emission of radiation from an excited state to a lower
electronic state of the same spin multiplicity (usually singlet states S1 to S0). This
process along with the other pathways deactivating the excited state can be
visualized in a Jablonski diagram (Figure 9).[16,91,93,95]
17
An organic molecule in solution is usually found in the lowest vibrational
level of the lowest electronic singlet state, S0, since the energetic bridge to higher
vibrational levels is typically larger than the thermal energy (E=kBT, where kb is
the Bolzmann constant and T is the temperature of the solution). This molecule is
then able to absorb a photon (of the correct energy) (Abs) to get excited to a
vibrationally excited level of a higher electronic singlet state (Sn with n ≥ 1)
(Figure 9). Subsequently, the vibrational excitation is quickly reduced due to
collisions with solvent molecules (i.e. vibrational relaxation, VR), rendering the
molecule at the lowest vibrational level of Sn. If this level is for example S2, the
molecule can undergo internal conversion (IC) to transform to an isoenergetic
excited vibronic state of S1. Hereafter, the molecule can once again rapidly relax
by vibrational relaxation to the vibrationless ground level of S1 (Figure 9).
So far, the processes discussed for the molecule to lose its excess energy are
too fast to allow competition from radiative processes, such as fluorescence.
However, internal conversion between S1 and S0 usually occurs on a longer
timescale since the energy gap between these levels is larger; giving radiative
processes a chance to take place. The competing processes from the vibrationless
state of S1 are internal conversion to S0 (IC, kIC), fluorescence (Fluo, kf) or
intersystem crossing to a triplet state T1 (ISC, kISC), where kn is the corresponding
rate constant for process n (Figure 9). Apart from the transformation of S1 to a
vibrationally excited isoenergetic level of S0 by internal conversion the molecule
can also emit a photon as fluorescence, thereby arriving at a vibrational level of
S0. The third option mentioned is intersystem crossing, in which the molecule in
the singlet excited state changes the spin of an electron, thereby altering spin
multiplicity, arriving at a vibrationally excited triplet state, T1 in Figure 9. Again
the molecule can undergo rapid vibrational relaxation to the ground vibrational
level of T1, after which a photon can be emitted as phosphorescence. Alternatively
the molecule can transform back to isoenergetic vibrational level of S0 through
intersystem crossing. In each of the above mentioned cases, the molecule will
quickly lose its excess vibrational energy left in S0 by vibrational relaxation.
A common measure for the fluorescence of a molecule is its fluorescence
quantum yield, defined as the ratio of the number of photons emitted per amount
of photons absorbed. This can also be understood as the probability of the excited
molecule in S1 to return to the ground state by emitting a photon amongst all
possible processes. Using the rate constants that were introduced above, this can
be written as
ߔ௙ ൌ
௞೑
( 12 )
௞೑ ା௞಺಴ ା௞಺ೄ಴
18
Frequently, fluorophores are also compared in terms of their brightness, which is
the product of their fluorescence quantum yield and absorption coefficient (usually
at the absorption maximum) (ߔ௙ ൈ ߝሺߣሻ).
Figure 9 Jablonski diagram giving a schematic overview of common radiative
processes a molecule can undergo (solid arrows) such as absorption (Abs),
fluorescence (Fluo) and Phosphorescence (Phos). Also the non-radiative
pathways are depicted by dashed arrows (vibrational relaxation, VR) or squiggly
arrows (Internal conversion, IC and Intersystem crossing, ISC).
Another important parameter often used to describe the photophysical
behavior of the molecule is its fluorescence lifetime, which is the average time the
molecule spends in the excited state following excitation. This can be described as
߬௙ ൌ
ଵ
( 13 )
௞೑ ା௞಺಴ ା௞಺ೄ಴
The fluorescence properties of the molecule can also be related to its absorbance
by means of the radiative rate constant, kf, and the absorption coefficient, ε(‫ݒ‬෤) and
can be expressed with the Strickler-Berg equation[96]:
݇௙ ൌ ʹǤͺͺͲ ൈ ͳͲିଽ ݊ଶ
௚೘
௚೙
ൻ‫ݒ‬෤௙ିଷ ൿ
ିଵ
‫ߝ ׬‬ሺ‫ݒ‬෤ሻ݈݀݊‫ݒ‬෤
( 14 )
where gm and gn are the degeneracies of the ground (m) and upper (n) electronic
state, n is the refractive index of the medium and ‫ݒ‬෤ f is the wavenumber of
emission. This equation is applicable for strong, broad-banded transitions in
molecules.
19
3.4.2 Steady state fluorescence
A common practice to study the photophysical properties of a chromophore in
bulk solution is to record an emission or excitation spectrum. This is done using a
spectrofluorimeter, for which the basic setup is shown in Figure 10. In the case
that an emission spectrum is to be obtained, a specific excitation wavelength will
be selected from the light source by the excitation monochromator which is then
focused onto the sample. Photons emitted by the sample are usually collected at
right angle to the incident radiation in order to reduce the interference of
unabsorbed excitation light. The emission monochromator is scanned through a
selected wavelength region, thereby yielding a spectrum of the emission intensity
per wavelength that reaches the detector. An excitation spectrum is measured in
the same manner; however, this time the emission monochromator remains parked
while scanning through the excitation energies.
Figure 10 Setup of a basic spectrofluorimeter with a source of UV/visible
radiation (for example a xenon arc lamp) that is led through a monochromator
selecting the correct radiation energy to excite the sample. The emission
monochromator then selects the correct wavelength of radiation to be observed
by the detector (usually at right angle of the excitation light).
As mentioned in the previous paragraph, fluorescence quantum yields are
frequently determined for molecules as a measure for their emissive capacity and
carries information about the rate constants depleting the electronically excited S1
state (Figure 9, equation 12). A common method to determine this value
experimentally is by using a reference substance with a known quantum yield,
ߔ௙ǡோ . This can be done in practice using the following relationship:
20
ߔ௙ ൌ ߔ௙ǡோ
ூ ஺ೃ ௡మ
( 15 )
మ
ூ ೃ ஺ ௡ೃ
where I is the integrated emission intensity, A is the absorbance at the excitation
wavelength and n is the refractive index of the solvent in which the
sample/reference is contained. R denotes the parameters corresponding to the
reference compound.
3.4.3 Steady state fluorescence anisotropy
As was described in section 3.1, the transition dipole moment for a transition
in a molecule from a ground electronic state m to an upper state n, exhibits a
magnitude and an orientation. While the main interest in the latter paragraph was
the magnitude, related to the probability of the transition occurring, the focus here
ሬԦ௡௠ . A molecule will preferentially absorb light that
will be on the orientation of Ɋ
has its electric field oscillating in the same direction as the transition dipole
moment. This implies that irradiating a sample with linear polarized light will
favour exciting those molecules with their transition moments aligned with the
polarization direction of the radiation. As fluorescence emission usually occurs
from the S1 to S0 electronic energy level, this transition is polarized parallel to the
S0-S1 absorption transition. Consequently, the emission from a molecule being
excited with polarized excitation light in the lowest energy absorption band (S0-S1)
will usually be co-linear, if it were not that a molecule in solution can reorient
during the time of excitation. In this way, the emission will be depolarized
depending on the time the molecule spends in the excited state as well as on its
size, shape and environment. To extract information about the molecule in this
way, the anisotropy, r, of a sample can be determined as follows:
‫ݎ‬ൌ
ூೇೇ ିீூೇಹ
( 16 )
ூೇೇ ାଶீூೇಹ
for which ‫ ܩ‬ൌ
ூಹೇ
ூಹಹ
is a correction factor and IXY is the intensity of polarized
emission upon excitation with polarized radiation. X and Y denote the polarization
directions of the excitation and emission light, respectively, i.e. horizontal (H) or
vertical (V).
If the molecule is immobilized on the time scale of fluorescence, the
anisotropy will contain information about the angle (ߙ௜ ሻ between the absorption
and emission moment of the ith transition. This can for example be achieved by
21
freezing the sample solution to a glass state. The anisotropy measured in this
special case is called fundamental anisotropy (‫ݎ‬଴ǡ௜ ሻ and is related to ߙ௜ by
ଵ
‫ݎ‬଴ǡ௜ ൌ ሺ͵ܿ‫ ݏ݋‬ଶ ߙ௜ െ ͳሻ
( 17 )
ହ
‫ݎ‬଴ can adopt values between -0.2 and 0.4, corresponding to an angle of 90° and 0°,
respectively, between the absorption and emission transition dipole moment. This
means that for most molecules a fundamental anisotropy value close to 0.4 is
expected upon excitation in the absorption region with a pure S0-S1 transition
(ߙ௜ ൌ Ͳι).
3.4.4 Time-resolved fluorescence
The lifetime of a fluorophore can be described as a function of the rate
constants depopulating the excited state, as was shown in equation 13. In practice
we can retrieve this parameter by studying the kinetics of the photophysical
process. If a population of molecules was excited with an infinitely sharp radiation
pulse, the initially excited population will decay over time to the ground state.
Since emission is a random process, it can be described by first order kinetics as
follows:
‫ܫ‬ሺ‫ݐ‬ሻ ൌ ‫ܫ‬଴ ‡š’ሺെ‫ݐ‬Ȁ߬௙ ሻ
( 18 )
where the excited state population of molecules remaining at time t can be
monitored by the fluorescence intensity ‫ܫ‬ሺ‫ݐ‬ሻ (which is proportional to it) and ‫ܫ‬଴ is
the initial fluorescence intensity at time 0.
Often multi-exponential expressions are observed for chromophores in
biological systems due to, for example, heterogeneity within the sample. In that
case the fluorescence decay can be described as
௝
‫ܫ‬ሺ‫ݐ‬ሻ ൌ σ௜ୀଵ ߙ௜ ‡š’ሺെ‫ݐ‬Ȁ߬௜ ሻ
( 19 )
for which αi is the amplitude of the ith lifetime (usually j ≤ 3 to allow a reliable fit)
and ∑ߙ௜ is normalized to unity. In that case an average lifetime is commonly used
to retrieve desired information. This can be described as the intensity-averaged
mean lifetime:
‫ ۄ߬ۃ‬ൌ
ೕ
σ೔సభ ఈ೔ ఛ೔ మ
( 20 )
ೕ
σ೔సభ ఈ೔ ఛ೔
22
or as the amplitude-weighted one:
‫ ۄ߬ۃ‬ൌ
ೕ
σ೔సభ ఈ೔ ఛ೔
( 21 )
ೕ
σ೔సభ ఈ೔
The choice of which average lifetime to use depends on the phenomenon being
studied. For Förster resonance energy transfer experiments the amplitudeweighted average (equation 21) is recommended.[16,95,97]
Figure 11 Schematic setup of a TCSPC measurement. A pulsed laser source
excites the sample, after which the emitted photons are recorded by a
microchannel plate photomultiplier tube (MCP PMT). The recorded data are fed
into a multichannel analyzer, resulting in a histogram.
In this thesis fluorescence lifetimes were determined with the technique timecorrelated single photon counting (TCSPC). The principle is based on the
proportionality between the probability of detecting a photon at a certain time after
excitation and the fluorescence intensity at this same time. A pulsed light source
(laser) is used to repeatedly excite the sample and the time between excitation and
the first photon reaching the detector is recorded (Figure 11). The time window is
divided into smaller intervals, so-called channels, and every counted photon is
recorded in the channel with the corresponding time interval. Measurements are
repeated until enough photons (often 10000 for statistical reasons) are counted in
the top channel. This finally yields a histogram with the amount of photons
counted per time interval (Figure 11), representing the fluorescence decay as a
function of time. To obtain a statistically correct measurement and to avoid two
photons originating from the same excitation pulse reaching the detector, the setup
is such that only every 100th pulse yields a photon arriving at the detector.
23
The lifetime can be retrieved by fitting the decay stored in the histogram
(R(t)) to an exponential function. Therefore, also the instrument response function
(IRF, E(t)) is recorded, which represents the way the laser pulse is seen by the
system. The IRF can then be convoluted with a theoretical decay (I(t-t’)) (in the
form of equation 19) to best fit the measured decay as follows:
௧
ܴሺ‫ݐ‬ሻ ൌ ‫ܧ‬ሺ‫ݐ‬ሻ ٔ ‫ܫ‬ሺ‫ݐ‬ሻ ൌ ‫׬‬଴ ‫ܧ‬ሺ‫ ݐ‬ᇱ ሻ‫ܫ‬ሺ‫ ݐ‬െ ‫ ݐ‬ᇱ ሻ݀‫ݐ‬Ԣ
( 22 )
where emission, i.e. I(t) starts at t=t’.
3.4.5 Förster resonance energy transfer (FRET)
Above (in section 3.4.1), the processes discussed for depopulating the excited
S1 state, concerned only ensembles of isolated molecules. However, molecules can
influence each other in different ways when coming in proximity. In case two
molecules are in each other’s vicinity (without overlap of the molecular orbitals)
and one is in an excited state, it can transfer its excitation energy to the second
molecule provided that this one has an absorption transition of matching energy,
known as the resonance condition. Not surprisingly, the molecules are usually
referred to as the energy donor and acceptor. If they are not too close by, their
interaction will mainly be coulombic in nature and can be described theoretically
considering point dipoles at the centers of the molecule. As a rule of thumb the
inter-chromophore distance is suggested to be at least four to five times the size of
the molecule. The phenomenon is referred to as resonance energy transfer (RET)
or Förster Resonance Energy Transfer (FRET) after Theodore Förster, who first
described the relationship between the rate constant for energy transfer and
spectroscopic observable properties[98,99]. Often the term Fluorescence Resonance
Energy Transfer (FRET) is also used, even though this name is somewhat
misleading since no emission of photons takes place from the donor molecule in
order for the energy to be relocated to the acceptor molecule.
According to Förster’s theory,[98,99] the resonance energy transfer rate, kET, is
given as follows
݇ா் ൌ
ଵ
బ
ఛ೑ǡವ
ோ
ቂ బቃ
଺
( 23 )
௥
଴
is the
where r is the distance between the donor and acceptor molecules, ߬௙ǡ஽
lifetime of the donor in the absence of the acceptor and R0 is the Förster radius,
given as
24
బ
ଽ଴଴଴ሺ௟௡ଵ଴ሻథ೑ǡವ
఑మ ௃
ܴ଴ ൌ ൤
ଵଶ଼ேಲ గఱ ௡ర
ଵȀ଺
൨
( 24 )
଴
is the fluorescence quantum yield of the donor without the acceptor
for which ߶௙ǡ஽
present, NA is Avogadro’s number, n is the refractive index of the medium, ߢ ଶ is
the orientation factor (see below) and ‫ ܬ‬is the overlap integral between the donor
emission ‫ܨ‬஽ ሺߣሻ (with the area under the curve normalized to unity) and the
acceptor absorption coefficient spectrum ߝ஺ ሺߣሻ, defined as
ஶ
‫ ܬ‬ൌ ‫׬‬଴ ‫ܨ‬஽ ሺߣሻߝ஺ ሺߣሻߣସ ݀ߣ
( 25 )
with
‫ܨ‬஽ ሺߣሻ ൌ
ூವ ሺఒሻ
( 26 )
ಮ
‫׬‬బ ூವ ሺఒሻௗఒ
ஶ
where ‫ܫ‬஽ ሺߣሻ is the recorded donor emission spectrum and ‫׬‬଴ ‫ܨ‬஽ ሺߣሻ݀ߣ ൌ ͳ.
The orientation factor (ߢ ଶ ) mentioned above relates the relative orientation of
the transition dipole moments of the donor and acceptor molecules to the rate of
energy transfer.ߢ ଶ is given as
ߢ ଶ ൌ ሺܿ‫ ߠݏ݋‬െ ͵ܿ‫ߚݏ݋ܿߙݏ݋‬ሻଶ
( 27 )
where θ defines the angle between both transition moments, whereas α and β
define the angle of the transition moment of the donor and acceptor, respectively,
with the vector connecting their centers. ߢ ଶ can adopt values between 0 for
transition moments at right angle and 4 for a collinear arrangement. The
orientation factor often introduces an uncertainty to ܴ଴ and hence to the
determination of r based on the recorded FRET efficiency (vide infra) since the
exact orientation of chromophores in complex systems is usually unknown. Many
dyes are externally attached to the system of interest by flexible linkers. In this
situation often a ߢ ଶ -value of 2/3 is assumed, which is valid in case of isotropic and
dynamic averaging of the orientations of the transition dipoles.[16,100]
The rate constant for energy transfer can now be used to obtain the efficiency
of energy transfer (߶ா் or EFRET) analogous to the fluorescence quantum
efficiency of a molecule (equation 12), by including the rate constants of all the
processes depleting the S1 state:
25
߶ா் ൌ ‫ܧ‬ிோா் ൌ
௞ಶ೅
௞೑ ା௞಺಴ ା௞಺ೄ಴ ା௞ಶ೅
ൌ
ோబ ల
( 28 )
ோబ ల ା௥ ల
Now it also becomes apparent that the Förster radius, ܴ଴ represents the distance
between the donor and acceptor molecule corresponding to a FRET efficiency of
50%. Since the FRET efficiency is dependent on the sixth power of the distance
between the donor and acceptor molecule it displays a sharp responsiveness to
distance changes in the range around R0. In this way FRET is often used as a
molecular ruler to obtain information about the relative separation between certain
parts of a macromolecule.[101,102]
In practice the FRET efficiency of a system can be determined by recording
the fluorescence quantum yield or lifetime of the donor with (߶௙ǡ஽ , ൏ ߬௙ǡ஽ ൐) and
଴
଴
without the acceptor present (߶௙ǡ஽
, ൏ ߬௙ǡ஽
൐ሻ (as can be derived from equation
28).
‫ܧ‬ிோா் ൌ ͳ െ
థ೑ǡವ
బ
థ೑ǡವ
ൌͳെ
ழఛ೑ǡವ வ
( 29 )
బ வ
ழఛ೑ǡವ
3.4.6 The tricyclic cytosine FRET-pair
The thought of retrieving structural information about DNA using FRET is a
tempting one, considering the well-ordered geometry of this molecule.[103] Often
externally tethered dyes attached by a linker are applied in quantitative FRET
measurements. This enables almost any commercially bright chromophore to be
implemented, allowing studies down to single-molecule level.[25,104,105] At the
same time the precise determination of the donor-acceptor distance (r) and
orientation factor ߢ ଶ is often complicated due to dipole diffusion and
reorientation.[53,56,57] Elegant studies have been reported, trying to achieve a better
control of dye position, resulting in FRET efficiency profiles responsive to the
orientation and distance change between the donor-acceptor pairs upon increasing
base-pair separation.[25,55,106,107] A first study involved a perylene acceptor attached
at the end of a stilbene-linked hairpin,[106] whereas other reports involved endstacked Cy3-Cy5 dyes attached by linkers[25,55] or base-surrogates consisting of a
more immobile pyrene-perylene FRET-pair[107].
In recent years our group has characterized and implemented a nucleobase
FRET-pair, consisting of the two tricyclic cytosine analogues, tCO and tCnitro
(Figure 12a).[40] The direct incorporation of these FBAs inside the DNA duplex
allows an almost precise control of their relative orientation on the time-scale of
26
energy transfer. The emission of the fluorescent donor, tCO, centered around 450
nm (in DNA),[39] overlaps the lowest energy absorption band (approximately 440
nm) of the non-emissive acceptor tCnitro[42] (Figure 12b). As for the earlier
characterized compound tC,[38] the emission of tCO is relatively insensitive to its
microenvironment,[39] in contrast to other FBAs. This results in an estimated
Förster radius of 27.2 Å when assuming a ߢ ଶ -value of 2/3, refractive index of 1.40
and an average value of the fluorescence quantum yield of tCO in dsDNA of
22%.[42] Consequently, this FRET-pair is more suited for studying short-range
structural changes compared to commercial dyes such as Cy3-Cy5 (R0~60 Å,
ߢ ଶ =2/3)[25]. As is to be expected for rigidly incorporated probes, increasing the
base-pair separation between both FBAs results in a FRET profile that closely
corresponds to that predicted for B-DNA (Figure 13).[40]
Figure 12 (a) tCO-tCnitro FRET-pair. The cytosine skeleton is grey. R is deoxyribose.
(b) Absorption/emission spectrum of tCO and tCnitro inside duplex DNA with the
overlap between tCO emission and tCnitro absorption highlighted in grey.
The main advantage of these stably incorporated probes is that they cause
only minor perturbations to the DNA duplex,[39,40] thereby allowing incorporation
close to or inside the site of interest. The FRET-pair can therefore be applied to
retrieve detailed structural information. In a recent study Preus et al. report on a
new methodology for the simulation and analysis of FRET in nucleic acids,
featuring a MATLAB based software, FRETmatrix.[108] The project implemented
the FBA FRET-pair to yield a 3D structure of kinked DNA. In this thesis,
however, the FRET-pair is rather implemented in yes/no types of questions, i.e. to
report on subtle local structural alterations in a specific DNA sequence upon
interactions with proteins.
27
Figure 13 FRET efficiency (EFRET) profiles for tCO-tCnitro FRET-pair as a function of
the number of base-pairs separating them (nDA). Efficiencies were determined
with time-resolved (EFRET <τ>) and steady state fluorescence (EFRET I). Also a fitted
profile based on B-DNA parameters is shown (Theoretical). (Data from our group
which were published in [40])
3.5 DNA melting
As was mentioned in 2.1, the two polynucleotide strands creating a DNA
double helix are held together by non-covalent forces. However, upon heating the
duplex, enough thermal energy will be provided at some point for both strands to
part, referred to as DNA melting. The temperature at which half of the duplexes
are denatured is called the melting temperature, which is dependent on the length,
sequence and concentration of the DNA as well as on its surroundings such as the
salt concentration and pH.
Melting temperatures are often reported as a measure of the stability of DNA
duplexes. They can be obtained by recording a UV/Vis melting curve, i.e. by
monitoring the change in absorbance at 260 nm as a function of the slowly rising
temperature. As the double helices melt, the absorbance will increase by
approximately 10-20 %. The decreased absorbance of the nucleobases inside
duplex DNA, called hypochromicity, can be explained by the close, more ordered
stacking of the bases in the duplex compared to single-stranded DNA. In this
sense the transition dipole moment of a stacked base will be shorter due to the
influence of the induced dipoles in the neighbouring bases.[109,110]
The melting temperature can be determined as the temperature at half
maximum of the curve or the temperature corresponding to the maximum of the
first derivative. In this thesis, the melting temperature is used mainly as a measure
28
of the (de)stabilization of the DNA caused by the incorporation of a modified
nucleobase.
Another way in which the melting process can be monitored is by using a
fluorophore whose emission is responsive to the duplex denaturation. As was
mentioned in section 2.2.2, the fluorescence of most FBAs is quenched in doublecompared to single-stranded DNA. In this way the rise in fluorescence of an FBA
can be monitored as the DNA duplex separates in response to the increasing
temperature. As for the absorption method, the melting temperature can be
determined as the maximum of the first derivative. The half maximum of the
signal can also be used, given that appropriate corrections are made for the
inherent temperature responsive change in fluorescence of the FBA. An advantage
of fluorescence melting, in contrast to UV melting is that it can also report on the
local stability of the DNA.
29
30
4 Results
In this chapter the main results contained in this work will be discussed. For a
complete overview of results and experimental procedures, the reader is referred
to the appended papers (I-VI).
4.1 Characterization of fluorescent nucleobase
analogues
The characterization of photophysical and base-mimicking properties of new
FBAs is essential for identifying applications they are suited for. Furthermore, this
can yield new insights into the correlation between their structure and
spectroscopic properties for future design of novel FBAs. The evaluation of
different aspects of a series of adenine analogues (papers I-III) and the cytosine
analogue tC (paper IV) are presented here.
4.1.1 Characterization of fluorescent adenine analogues
To increase the diversity of FBAs with improved structural and spectroscopic
properties, two fluorescent adenine analogues, triazole adenine (AT) (paper I,
Figure 14) and quadracyclic adenine (qA) (paper III, Figure 14) were synthesized.
Their photophysical properties, both as nucleosides and inside DNA, are discussed
here.
Figure 14 Nucleosides of AT, qA and the ethyl esters of the new A7T-family.
31
As a follow-up study, trying to overcome the DNA duplex destabilizing effect
of AT (vide infra), a new family of 7-modified triazole adenines (A7T-family,
Figure 14) was developed, of which the photophysical characterization of the
monomeric form is also presented (paper II).
Spectroscopic characterization of the monomers
Before FBAs are incorporated into DNA, it is important to understand their
photophysical properties as monomers. The spectral features of the adenine
analogues qA, AT and the A7T-family (Figure 14) are presented in Figure 15 and
Table 1. The absorption of all compounds is red-shifted compared to adenine
(centered around 260 nm), allowing for selective excitation due to an absorption
tail extending further than that of the natural nucleobases. Quadracyclic adenine
(qA) has a well resolved lowest energy absorption band, located at 335 nm (in
water), whereas triazole adenine (AT) has its lowest energy absorption centered
around 282 nm in water and therefore not fully separated from the natural bases.
The A7T-family (280-293 nm in methanol) absorbs in a similar region as AT (286
nm in methanol) as well as a series of previously characterized C8-modified
triazole adenines (289-296 nm in THF).[37] AT shows similar absorption envelopes
in water and methanol, which is also observed for qA (Figure 15a). The A7Tfamily was studied in methanol due to a poor solubility in water (20b and 20c
were also studied in ACN or DCM, vide infra). Within this group of analogues, a
more structured absorption can be observed for the compounds containing a
nitrogen (Figure 15e) instead of carbon (Figure 15c) on the 8-position.
The shape and position of the emission spectra of all triazole adenines (AT
and the A7T-family) are very similar in methanol, with exception of compound
20b (Figure 15d and f). The latter shows an extensive Stokes shift of ~12800 cm-1
(∆λ=175 nm) compared to the other compounds in its series. This is due to a more
red-shifted emission maximum (468 nm) in methanol, which may be caused by
intramolecular charge transfer (ICT) in the excited state from the electron
donating amino group. The shift of its emission maximum to shorter wavelengths
in dichloromethane (DCM, apolar and aprotic) (λem,max=393 nm) and in acetonitrile
(ACN, polar and aprotic) (λem,max=428 nm) are in accordance with a possible
excited ICT state, which may be stabilized by polar protic solvents (Table 1,
Figure 15f).[111,112] As a comparison, the related conjugated analogue 20c was also
investigated in DCM and showed no changes in emission maximum. Furthermore,
no significant changes could be detected in the absorption spectrum of both 20b
and 20c in DCM, methanol or ACN (only for 20b).
qA shows a lower emission energy than the triazole adenines (except for 20b
in methanol) both in water (456 nm) and methanol (440 nm) with a more
structured spectral envelope detected in methanol (Figure 15b).
32
Figure 15 Normalized absorption (a) and emission spectra (b) of AT and qA in
water and methanol (MeOH). Also absorption (c, e) and emission (for Φf>0.2%)
(d, f) of members of the A7T-family with (c, d) and without nitrogen (d, f) on
position 8 of adenine. Also AT in MeOH is included for comparison as well as
emission of 20b (f) in DCM (pink, dotted) and ACN (pink, dashed).
AT is highly fluorescent both in water (Φf=61%) and in methanol (Φf=49%)
(Table 1). Its quantum yield is similar to 2-aminopurine (2-AP) (68%)[32] as well
as the previously reported 8-modified adenine analogue, 8-vinyl-deoxyadenosine
(8vdA) (66%)[69] in aqueous solution and exceeds the quantum yields of the
pteridine adenine analogues 6MAP (39%) and DMAP (48%)[61]. The A7T- family,
on the other hand, is significantly less emissive than AT in methanol with
compound 12a, a sister compound of AT, exhibiting the highest quantum yield
(Φf=5.2%) of the series (Table 1). The second sister compound, 11a, lacking the
nitrogen on the 8-position, has a much lower quantum yield (Φf=0.75%). Within
each group of the A7T-compounds, based on a nitrogen or carbon being
incorporated on the 8-position, all other compounds show lower quantum yields
33
than the two sister compounds of AT in methanol. As was also observed for a
previous class of C8-triazole adenines, the compounds with extended aromatic
systems (19, 20a, 20b and 20c) are the least fluorescent within their groups.[37]
These compounds also show absorption spectra that extend to longer wavelengths
and have the highest extinction coefficients, which could be explained by their
more extended conjugated system.
Table 1 Fluorescence quantum yield (Φf), lowest energy absorption maximum
(λA,max), emission maximum (λEm,max) and corresponding extinction coefficient
(εmax) at λA,max of each adenine analogue (Cpd) dissolved in water, methanol
(MeOH), acetonitrile (ACN) or dichloromethane (DCM). Also the brightness
(∝Φfൈ εmax) is shown.
Φf
λA,max
λEm,max
εmax
Solvent Φf ൈ εmax
-1
-1
(nm)
(nm)
(%)
(M cm )
(M-1cm-1)
61
282
353
16500c
Water
10065
AT
49
286
346
16500
MeOH
8085
0.75
282
369
11200
MeOH
84
11a
0.39
281
366
11800
MeOH
46
11b
5.2
290
343
10700
MeOH
556
12a
4.4
290
340
10800
MeOH
475
12b
b
<0.1
280
14500
MeOH
<15
19
<0.2
288
345
13600
MeOH
<27
20a
0.62
293
468
14000
MeOH
87
c
24
295
428
14000
ACN
3360
20b
21
297
393
14000c
DCM
2800
a
c
<0.1
293
359
15300
MeOH
<15
20c
<0.1
295a
359
15300c
DCM
<15
6.8
335
456
5000
Water
340
qA
10.6
335
440
6300
MeOH
668
a
Absorption maximum (λA,max) estimated for lowest energy absorption shoulder
b
Emission maximum omitted due to the weak fluorescence. cExtinction
coefficient assumed to be the same as in methanol.
Cpd
Interestingly, compound 20b shows a dramatic increase in quantum yield in
ACN (Φf~24%) and DCM (Φf~21%) compared to in methanol (Φf=0.62%),
indicating efficient quenching by protic solvents (Table 1). This could be due to a
more efficient internal conversion in case of hydrogen bonding with the
solvent.[111] As a comparison, the quantum yield of 20c in DCM remained
unaltered compared to in methanol (Table 1). A similar quenching behavior was
also observed for the proposed twisted intramolecular charge transfer-state of
N6,N6-dimethyladenosine.[113]
Finally, qA shows a lower emission than AT, with a moderate quantum yield
of 6.8 % in water. The extinction coefficient corresponding to its lowest energy
34
absorption maximum (5000 M-1cm-1) is reduced approximately three-fold in
comparison with AT (16500 M-1cm-1).
In general, AT is by far the brightest (∝Φfൈε) of all ten adenine analogues
discussed in this work, both in water and methanol (Table 1). Its brightness in
water (~10000 M-1cm-1) is also more than double as high as for 2-AP
(~4000 M-1cm-1)[32,114].
Characterization of the lowest energy absorption band of AT and qA
To get more insight into the different transitions contained in the lowest
energy absorption band of AT and qA, their fundamental anisotropy was recorded
in a vitrified matrix (Figure 16). This information is important in case the
analogues are to be applied in anisotropy- or FRET-measurements (vide infra,
anisotropy of AT inside DNA in high viscosity sucrose solutions). Furthermore,
this allows for comparison of the energy of transitions with predictions from
quantum chemical calculations.
Figure 16 Excitation anisotropy spectrum of AT (a) and qA (b) in a vitrified
H2O/ethylene glycol matrix (1:2 mixture) at -100°C (grey line). Also the isotropic
absorption is shown (black) as well as the relative intensities predicted for the 3
lowest energy transitions (grey bars) for AT (methyl instead of pentyl substituent
on triazole ring, left to right f=0.094; 0.059 and 0.72)[37] and qA (left to right
f=0.1049; 0.1968 and 0.0594; TDDFT calculations).
For AT, a virtually constant value of r0=0.38 was recorded between 280 and
320 nm, indicating a single electronic transition dominating this region (Figure
16a). A strong and isolated lowest energy transition (f=0.72) was predicted at 315
nm for a related methyl substituted C8-triazole adenine (calculations from a
previous study)[37]. This prediction corresponds reasonably well to our findings for
AT, even though the transition is slightly shifted. Furthermore, the value of r0 is
close to the theoretical maximum (0.4) for these excitation energies, suggesting
virtually parallel absorption and emission transition moments.
35
The fundamental anisotropy spectrum for qA shows three main plateaus
between 280 and 450 nm (Figure 16b). The first one, associated with the lowest
energy absorption transition is found between 390 and 450 nm with r0=0.36. As
for AT, this value is close to the theoretical maximum, suggesting a single
electronic transition in this region, polarized parallel to the emission transition
moment. A second plateau is located around the absorption maximum (~330 nm,
r0=0.25) and a third one around 300 nm (r0=0.05), where also a shoulder can be
observed in the absorption spectrum. These findings suggest three absorption
transitions between 280 and 450 nm, with the lowest energy transition around 400
nm (S0-S1). The positions as well as relative intensities of these three transitions
seem to correspond well to theoretical predictions, albeit somewhat shifted (Figure
16b). Based on the second plateau (~330 nm, r0=0.25), the angle between the
transitions from S0 to S1 and from S0 to S2 can be estimated to be ~30°. The third
plateau (~300 nm, r0=0.05) suggests the angle between S0-S1 and S0-S3 to be at
least 50°. This is in line with a larger angle predicted between S0-S1 and S0-S3
(83°) than between S0-S1 and S0-S2 (15°) by calculations.
The predicted oscillator strength for the lowest energy transition of qA
(f=0.0594) is approximately 12 times lower than that predicted for the ATderivative (f=0.72)[37]. Knowledge of the lifetimes (τf,AT=1.8 ns and τf,qA=3.2 ns;
water) and quantum yields of the analogues (Φf,AT=61% and Φf,qA=6.8%; water)
also allows the calculation of their radiative rate constants, kf, which for qA is
roughly 16 times decreased compared to AT. This corresponds reasonably well to
the predicted decrease in the oscillator strength as kf and f are related (see equation
9 and 14).
Structure and stability of DNA duplexes containing AT and qA
In order to investigate the influence of AT (paper I) and qA (paper III) on the
DNA duplex integrity, they were incorporated in a series of decamers for which
the bases directly flanking the analogues were varied (Table 2). In this way the
effect of as many combinations of purines and pyrimidines as possible was
evaluated, as immediate 5’ and 3’ neighbors, depending on the amount of
compound available. For AT, ten different combinations of flanking bases were
tested, whereas a larger amount of qA allowed us to investigate twelve sequences.
This also included a decamer holding two qA molecules (CT,TA).
The melting temperature of all AT and qA-modified duplexes was recorded
and compared to the corresponding natural ones in order to evaluate the effect of
the adenine modifications on duplex stability (Table 2). AT causes an average drop
in melting temperature of 8°C, indicating a moderate destabilization of the DNA
36
duplexes, which is most pronounced for sequences in which AT is flanked by a 3’
purine. This destabilizing effect is probably due to the triazole group and pentyl
chain incorporated on C8 of adenine. Steric clashes of these substituents with the
DNA backbone would account for a significant energy cost in case AT was to be
accommodated in the duplex in its natural anti-conformation. Even though smaller
C8-substituents on purines are only insignificantly perturbing the duplex
integrity,[69,115] more bulky modifications have been observed to be
destabilizing,[116-118] similar to AT. For some C8-modified purines, a preference for
the syn conformation has been reported,[119,120] whereas a syn/anti equilibrium has
been suggested for others inside DNA.[121] Also for AT, the C8-triazole
modification may cause a stabilization of the syn-conformation, relative to the
anti-orientation.
Table 2 DNA melting temperatures of different 10-mer duplexes containing AT
T
(TmA , left) or qA (TmqA, right). Sequences are named after the bases directly
neighboring AT or qA. Also the difference in melting temperature (∆Tm) is shown
between the modified and unmodified (containing natural adenine) DNA duplex.
T
Neighbouring
TmqA
∆Tm
a
bases qA
(°C)b
(°C)
46.3
-0.3
AA
48.3
0.0
GA
51.9
0.4
GG
GC
52.3
1.5
AC
52.7
1.9
TA
48.5
3.2
CA
53.2
3.3
TG
48.7
3.4
CC
58.2
5.3
CT
56.1
6.5
TT
50.6
6.9
CT,TA
55.6
9.2
a
Full sequence is 5’-d(CGCAXAYTCG)-3’, with exception of sequence CT,TA; 5’d(CGCATAATCG)-3’ and TT, 5’-d(CGATTATGCG)-3’. A represents AT or qA and the
bases flanking them, X and Y, give the sequence name XY. Purines are shown in
bold and pyrimidines in italic. bSamples were prepared in phosphate buffer (pH
7.5, 500 mM (AT) or 200 mM (qA) Na+) at duplex concentrations of 2.5 μM.
Neighbouring
bases AT a
CA
GG
AA
TA
TG
GA
CT
AC
CC
GC
TmA
(qqC)b
43
45
41
39
42
43
45
46
49
49
∆Tm
(qqC)
-9
-9
-9
-8
-8
-7
-7
-7
-7
-5
In order to investigate how stable AT is stacked into the DNA, its anisotropy
was recorded inside duplexes AA and TA in viscous sucrose solutions. This high
viscosity hinders the motion of the DNA helices on the time-scale of fluorescence,
resulting in a limiting anisotropy value of 0.34 reached by AT. The decrease
compared to r0 of 0.38 for the monomer in a vitrified matrix was predicted to be
37
due to an internal wobble of 16° inside the base-stack, compared to ~5° for
canonical bases and 21° for ethidium bromide (personal communication from
S.Preus).[122] Consequently, despite a possible conformational change around the
glycosidic bond, AT seems to be stacked more stably in the duplex than the
intercalating dye ethidium bromide, indicating that it still has base-pairing
capacity to thymine (vide infra).
In contrast to AT and other adenine analogues such as 2-AP,[33] 6MAP and
DMAP,[61] qA generally stabilizes the DNA duplexes (Table 2). This can be seen
by an average increase in melting temperature of 3°C compared to the natural
duplexes. The largest stabilization is seen when qA is flanked by a pyrimidine on
the 5’ side, whereas this effect is minimal for a 5’ neighboring purine. Sequences
AA, GA and GG, having two purines neighboring qA, are virtually equally stable
as the corresponding duplexes containing a natural adenine instead. The overall
stabilization caused by qA is probably due to better π-π stacking with its
neighboring bases due to the extended ring system. Similar observations have
been made for the tricyclic cytosines tC[41] and tCO[39], which also showed the
largest stabilizing effect in case they had a 5’ pyrimidine neighbor.
The two ring extension for qA, on position 6 and 7 of adenine, is probably
well accommodated in the major groove. Also other purines modified on the 7position have shown a stabilizing effect on DNA duplexes.[123-125] Overall, qA
seems to be stably incorporated into the DNA duplex. Moreover, fluorescence
melting curves for duplex AA (Figure 17) yielded a virtually identical melting
temperature (46.8°C) to that recorded with UV/Vis melting (46.3°C), indicating
no pre-melting around qA.
Figure 17 Fluorescence melting curve of qA inside sequence AA in phosphate
buffer (pH 7.5, 200 mM Na+, duplex concentration of 2.5 μM).
In order to evaluate the effect of AT and qA on the overall DNA
conformation, the CD spectra of all modified sequences were recorded (Figure
38
18). Both AT- and qA-modified duplexes show a general signature of regular Bform DNA, for which the typical parameters were described in Section 3.3.
However, for AT, some heterogeneity was observed in the shape of their CD
spectra between sequences compared to the corresponding unmodified duplexes
(Figure 18a). There are sequences with an almost identical CD-profile to the
natural duplexes (CT and AA) and others which show slight differences
corresponding to the AT absorption bands (CA, TG and CC). Moreover, more
distinct differences were observed for sequences TA, AC, GA, GC and GG, of
which the latter three show the most perturbed CD spectra. As an example, the CD
spectrum of GG with AT or adenine is shown in Figure 18c, as well as the
difference in absorption between both duplexes. As can be seen, the deviations in
the CD-signature between both samples correspond well to the difference in
absorption.
Figure 18 CD spectra of DNA duplexes containing AT (a) and qA (b) (sequences in
Table 2; CT,TA and TT are unique for qA). No CD signal is seen for qA above 350
nm. Also the comparison of CD signatures of sequence GG containing AT (c) and
qA (d) (black) to the spectrum of the unmodified GG duplex (grey) is shown. For
AT, the difference in the corresponding absorption spectra is shown as well (AmodAunmod, dotted). M-1 on the x-axis refers to the duplex concentration (2.5 μM).
Contrary to AT, the CD spectra of all samples containing qA are very similar
in shape (Figure 18b). Furthermore, they all correspond well to those of the
unmodified sequences. As for AT, sequence GG is shown as an example ( Figure
18d). This finding may be expected since qA seems to be better tolerated inside
39
DNA than AT according to the melting studies. Additionally, also for sequence
CT,TA containing two qA molecules, a similar CD-signature was observed,
indicating that also the incorporation of several qA molecules preserves the BDNA structure. As for some AT-modified duplexes (CT and AA), no specific CD
signal can be detected corresponding to the lowest energy absorption band of qA,
centered around 337 nm. Similar observations have been made for the cytosine
analogue tCO,[39] for which we do not yet have a satisfactory explanation.
Base-pairing preference of AT and qA
The base-pairing specificity of AT and qA was investigated by recording the
melting temperature for three different duplexes in which the analogues were
placed opposite of a guanine, cytosine or adenine (Table 3). Sequences were
chosen so that the effects of neighboring purines (GA), pyrimidines (CT) and a
combination of both (CA for AT and TA for qA) could be evaluated.
An additional decrease in melting temperature of ~8°C was observed for AT
opposite of cytosine or guanine in comparison to the AT-thymine case, yielding a
total destabilization of approximately 16°C. This value is comparable to an
adenine-adenine mismatch in these sequences causing a decrease in melting
temperature of 14°C. It is therefore plausible that AT still exhibits hydrogen
bonding capacity to thymine despite its destabilizing effect. Surprisingly, no
additional decrease in melting temperature was observed when placing AT
opposite of adenine compared to the AT.T case indicating that AT can form equally
stable base-pairs with thymine as with adenine. A potential AT.A base pair is
presented in Figure 19 and compared to a possible AT.T pair with AT in the syn
conformation around the glycosidic bond. Perturbations in the base-pairing
patterns are not unexpected upon adding (often bulky) substituents to the natural
bases. Also other FBAs such as 2-AP[34-36] and BPP[58] have been shown to have
deviating base-pairing specificities compared to their natural counterparts.
Figure 19 Putative AT.A (left) and AT.T (right) base-pair with AT in the syn
conformation around the glycosidic bond. R1, R2 and R3 represent the rest of the
DNA structure. The triazole substituent on AT is shown in grey.
40
In the case of qA, a significant destabilization was observed for a mismatch
with cytosine, guanine and adenine compared to the qA-thymine case, indicating
that qA shows specific base-pairing to thymine (Table 3). The decrease in melting
temperature was the lowest for qA opposite of guanine and the same observation
was made for the mismatched sequences with natural adenine. These findings
indicate that the base-pairing pattern of qA is unperturbed compared to natural
adenine. This is also in line with the assumption that the extended rings are well
accommodated in the major groove, leaving qA positioned similar to natural
adenine in the base-stack.
T
Table 3 Melting temperatures of sequences containing AT (TmA -X) or qA (TmqA-X)
mismatched with G,C or A are presented as well as the difference with the
T
melting temperature (∆TmA and ∆TmqA) of the corresponding matching duplex
containing the analogues opposite of T.
Neighboring
basesa
Xc
T-X
TmA
(qqC)d
T
∆TmA (qC)
TmqA-X (oC)d
∆TmqA (oC)
G
36
-7
38.8
-9.5
C
36
-7
34.1
-14.2
A
43
0
34.2
-14.1
G
36
-9
47.6
-8.5
CT
C
36
-9
39.3
-16.8
A
46
1
40.5
-15.6
G
36
-7
37.0
-11.5
CA (AT)b
C
36
-7
32.8
-15.8
TA (qA)
A
42
-1
33.5
-15.0
a
b
T
Full sequences in Table 2. Sequence CA was used for A and TA in the case of
qA. c Base opposite of AT or qA. dConditions were similar as described for Table 2.
GA
Photophysical properties of AT and qA in DNA
The fluorescence properties (Φf and λEm,max) of AT and qA inside DNA are
summarized in Table 4 and Table 5, respectively. As is common upon
incorporating FBAs into nucleic acid systems,[27,29,30,32,69,73,115] the quantum yield
of both AT and qA generally is significantly decreased in single- and doublestranded DNA. Furthermore, variations can be seen between neighboring bases,
which means that the emission of both probes is responsive to their immediate
surroundings.
For AT (Φf=61% in water), the average quantum yield for all combinations of
flanking bases is decreased approximately 12 times for the single strands and
additionally 5 fold in double-stranded DNA. Furthermore, large variations can be
seen depending on the neighboring bases. These observations may be due to
different levels of stacking with the surrounding bases in single- and double41
stranded DNA as well as different conformations of AT inside DNA. In general the
highest quantum yields are detected in the case where AT is flanked by an adenine
on the 3’-side (exception duplex GA), whereas the lowest quantum yields are
detected for a neighboring 3’ guanine. The fact that sequence GG is among those
with the lowest quantum yield may be due to photoinduced electron transfer
(PET). Guanine has the lowest oxidation potential among the natural bases and
could therefore quench the excited state of the neighboring AT by PET. Decreases
in quantum yield due to neighboring guanines have also been observed for other
FBAs such as 2-AP[126,127], tCO[39] and BPP[58]. In contrast to sequences GG and
TG, those with a guanine as the 5’-neighbor of AT (GA and GC) have higher
quantum yields in single- and double- stranded DNA. Similar observations had
been made for tCO previously,[39] indicating that not only proximity to a guanine
but also stacking and relative orientation are important parameters for PET.
Table 4 Fluorescence quantum yields (Φf) of the AT-modified single-and doublestranded oligonucleotides, Φf,SS and Φf,DS, respectively. Also their emission
maxima are shown (λEm,max).
Neighboring
Φf,SS
λEm,max
Neighboring Φf,DS
λEm,max
basesa
(%)
(nm)
basesa
(%)
(nm)
21
353
5.0
350
AA
AA
CA
9.0
351
TA
1.6
353
TA
7.6
352
CA
1.1
354
GC
4.2
358
AC
0.9
349
AC
3.1
357
0.8
351
GA
CC
2.1
355
CT
0.5
351
1.8
353
GC
0.5
351
GA
CT
1.4
358
CC
0.4
352
TG
0.6
355
TG
0.3
351
0.5
355
0.3
354
GG
GG
a
Sequences are named as described under Table 2.
A decrease in average fluorescence quantum yield of ~4 and additionally ~5
times can be detected for qA incorporated into single-stranded and duplex DNA,
respectively, compared to the monomer in water (Фf,λEx337nm=6.8%) (Table 5). As for
AT, this may be explained by various levels of stacking with the surrounding
bases. In the case of qA, the highest quantum yields are detected when it is
positioned between two purines, including sequence GG, which is in contrast to
findings for AT and the FBAs mentioned above. However, we cannot draw any
conclusions regarding PET since we have not determined the redox potentials of
qA. As for AT and other FBAs, the quenching of qA inside DNA is probably due
to a combination of mechanisms, such as base stacking and collisions with the
neighboring bases,[78] hypochromism[39,128] as well as PET,[127] which are
dependent on the surrounding sequence and structural parameters.
42
Maximum quantum yields for AT (21%) and qA (5.8%) in single-stranded
DNA exceed values reported for nucleic acids containing 2-AP.[32] For AT, this
value is approximately 10 times higher compared to 2-AP[32] and roughly five and
two times higher than values reported for 6MAP and DMAP[61], respectively.
Inside duplexes, the maximum quantum yield of AT (5%) and qA (0.6%) is
roughly five times higher or of the same order, respectively, as 2-AP.[32] The
overall brightness (∝ <Φf,DNA> ൈ εmax) of AT inside duplex DNA (~190 M-1cm-1)
is, thus, roughly 10 times higher than qA (~20 M-1cm-1) and 3 times higher than 2AP (~60 M-1cm-1) in a similar environment.[32,114]
Table 5 Fluorescence quantum yields (Φf) of the qA-modified single- and doublestranded oligonucleotides, Φf,SS and Φf,DS respectively. Also their emission
maxima are shown (λEm,max).
Neighboring
Фf,SS
λEm,max
Neighboring Фf,DS
λEm,max
bases[a]
(%)b
(nm)[b]
bases[a]
(nm)[b
(%)b
5.8
432
0.6
434
AA
AA
5.5
433
0.6
435
GG
GG
4.4
432
GC
0.6
453
GA
AC
2.0
434
0.5
438
GA
GC
1.6
438
CA
0.3
444
CA
1.0
433
CT,TA
0.3
457
TA
0.6
434
AC
0.2
441
CT,TA
0.4
439
TA
0.2
451
TG
0.4
437
TG
0.2
453
CC
0.4
440
CT
0.2
459
CT
0.3
443
CC
0.2
452
TT
0.2
434
TT
0.2
452
a
Sequences are named as described under Table 2.
Conclusion about AT, qA and A7T-family
The emission of both qA and AT is sensitive to their micro-environment. As
for other FBAs, this sensitivity could be applied in studies concerning dynamics,
structure[79,80,129] and interactions of DNA with other molecules[82,84,130]. Further,
qA is a better adenine isomorph, whereas AT is the better choice in case a higher
brightness is crucial.
Even though the members of the new A7T-family are dramatically less
emissive than AT in methanol, it will be interesting to investigate the most
promising compounds, 12a (Φf=5.2%) and 12b (Φf=4.4%), in DNA since it is
very likely that they will be well accommodated in the DNA duplex. Also 20b
(Φf,DCM=21%, Φf,ACN=24%) would be interesting to investigate in a DNA context
where it is more shielded from protic solvent molecules that efficiently quench its
fluorescence (Φf,MeOH=0.62%). However, one should consider that also inside
43
DNA there will be effects of hydrogen bonding from the opposite base and the
triazole extension exposed into the major groove.
4.1.2 Photostability of the tricyclic cytosine analogue tC
In order to develop FBAs which are brighter and more photostable to allow
single-molecule studies, a good understanding of their photodegradation is
important. Therefore, the characterization of the photostability of tC, a member of
the tricyclic cytosine family, is presented here (paper IV) (Figure 20a). The
photoconversion of tC to its photoproduct tC# was studied both as a monomer and
inside duplex DNA.
Upon intense irradiation of the tC monomer in its lowest energy absorption
band (~375 nm) the latter is slowly replaced by a blue-shifted lowest energy
absorption band centered around 310-320 nm (Figure 20b). The two isosbestic
points (at 285 nm and 343 nm) indicate the conversion of tC to a single
photoproduct tC#, with a strongly decreased blue-shifted emission (Φf<2%)
compared to tC. Analysis of the photoproduct by NMR spectroscopy and mass
spectrometry as well as the increase in the photoconversion rate upon saturation
with O2 suggest tC# to be the sulfoxide form of tC (Figure 20a). Furthermore, the
lowest energy electronic transition of tC# predicted by DFT-calculations (317 nm)
corresponds well to the observed absorption band of the photoproduct in water.
The sulfoxide could be formed after tC undergoes intersystem crossing from its
first excited singlet state to a triplet state and reacts with O2.
Figure 20 (a) Tricyclic cytosine analogue tC and its suggested photoproduct tC#.
(b) Evolution of the UV-Vis absorption spectrum of the potassium salt of tC in
water upon irradiation at 375 nm with a 150 W Xe lamp. The spectra were
recorded over an irradiation period of 10 hours.
44
The photostability of tC was not only studied as a monomer, but also inside
duplex DNA, where it was incorporated into a 10-mer oligonucleotide flanked by
two adenines (Table 6, Figure 21). The photoconversion of tC inside doublestranded DNA was found to be significantly slower under similar conditions as for
the monomeric form. This also supports the hypothesis that oxidation of the
sulfide leads to the photoproduct, since tC would be less accessible to O2 inside
the base stack.
Table 6 Melting temperatures (Tm) of DNA duplexes containing a central cytosine,
tC or its photoproduct tC#.
Sequencea
Xb
Tm (°C)
5’-d(CGCAACATCG)-3’
G
41.6
5’-d(CGCAAtCATCG)-3’
G
43.6
#
5’-d(CGCAAtC ATCG)-3’
G
26.0
5’-d(CGCAAtCATCG)-3’
T
24.0
a
Complementary sequence is 5’-d(CGATXTTGCG)-3’ bX is the base in the
complementary sequence that is opposite of the central base C, tC or tC#.
Interestingly, the stability of the DNA is significantly decreased upon
photoconversion of tC (Tm=43.6°C) to tC# (Tm=26.0°C), resulting in partly
denatured duplexes at room temperature (Table 6). This destabilization is
comparable to a tC-thymine mismatch (Tm=24°C) and could be explained by steric
effects of the oxygen atom. The latter is predicted to be perpendicular to the
surface of tC# (DFT calculations) and would therefore interfere with the base
flanking tC# on the 5’ side.
Figure 21 (a) Evolution of the UV-Vis absorption spectrum of tC inside duplexDNA during fast photolysis upon irradiation with a Nd:YAG laser at 420 nm for 012 minutes (7 ns pulses, 10 Hz, 2.1 mJ/pulse). The solution was bubbled with O 2
every 2-4 minutes. Spectra are smoothed by adjacent averaging (10 points). (b)
CD-spectra of DNA duplexes containing a central natural cytosine, tC or tC#.
Sequences are shown in Table 6.
45
On the other hand, it should be mentioned that CD-spectra for the DNA
duplexes containing tC# opposite of guanine or a tC-thymine mismatch still show
the same spectral features of B-DNA (see 3.3) as the tC-G case and the natural
duplex (Figure 21b). Therefore we cannot exclude the possibility that the observed
destabilization is (in part) due to a reduced base-pairing capacity of tC# to G.
Finally, when taking a closer look at the recorded CD spectra, the tC#-profile
looks more similar to the unmodified duplex (C) than that of the tC-modified
duplex. This could be explained by the difference in their isotropic absorption
(Figure 20b).
As was mentioned in the paragraph above, the photodegradation of tC inside
DNA was much slower (12-24 hours, 200 W Hg lamp) under similar conditions as
compared to the monomer (~4 hours). However, fast photolysis (<10 min) can be
achieved using more intense irradiation (Figure 21a). This indicates the potential
of the photoconversion of tC as a trigger for DNA melting with potential
applications in photoinduced therapeutics and in functional DNA-based
nanodevices.[131-133]
4.2 Application of the tricyclic cytosine FRETpair to investigate protein-DNA interactions
The mitochondrial DNA (mtDNA) is a circular molecule that encodes 22
tRNAs (transfer RNA), 2 rRNAs (ribosomal RNA) and 13 proteins involved in the
respiratory chain. Transcription is initiated from two sites, i.e. the light- and
heavy-strand promoters (LSP and HSP1).[134] A third transcription site has been
located as well, HSP2, of which the function and sequence requirements still
remain to be established.[135,136]
In mammalian cells, transcription involves the mitochondrial RNA
polymerase (POLRMT, human or Polrmt, mouse) and transcription factor B2
(TFB2M, human and Tfb2m, mouse). Also a third factor, the transcription factor
A (TFAM, human or Tfam, mouse), has been reported to be involved,[136-138] even
though its role has been debated lately[139]. In paper V the role of TFAM in
mammalian transcription is established, whereas paper VI focuses on the assembly
of the transcription machinery to the promoter. The nucleobase FRET-pair,
consisting of tCO and tCnitro (see 3.4.6)[40], was applied to monitor structural
changes around the transcription start site in both studies.
46
Structural influence of TFAM on the heavy strand promoter
TFAM binds site-specifically to a site upstream of the transcription start site
of LSP and HSP1 (Table 7).[140,141] However, it is also known to bind nonsequence specifically as an mtDNA packaging factor, which wraps and bends the
DNA.[142-145] Even though a recent report disputed the role of TFAM in
mammalian mitochondrial transcription,[139] in vitro transcription experiments
presented in paper V establish the necessity of TFAM. Only at lower salt
concentrations (mainly for HSP1) or on a negative supercoiled template (only for
HSP1), transcription can be detected in the absence of TFAM. This observation
may indicate that the role of TFAM in transcription initiation is to induce negative
supercoils in promoter DNA, thereby stimulating local melting by POLRMT and
TFB2M. Furthermore, the TFAM-binding site is essential, since transcription from
both LSP and HSP1 is abolished for a template with mutations in the specific
TFAM-binding site.
The tCO-tCnitro FRET-pair was applied to investigate whether TFAM induces
structural changes around the transcription start site. The tCO donor and tCnitro
acceptor were therefore incorporated close to the HSP1 transcription start site with
6bp distances separating them in the opposite strands (Table 7). Since the probes
are rigidly incorporated in the DNA duplex, their FRET efficiency is expected to
respond to subtle structural changes around the site of incorporation.
Table 7 DNA sequence of the heavy strand promoter (HSP1), in which the FRET
donor tCO and acceptor tCnitro were incorporated at sites X1-X7 (see Table 8). The
TFAM binding site is underlined and the transcription start site is highlighted
grey. Mutations in the sequence are italic.
DNA sequencea
5’-d(-----CACACACCGCTGCTAACCCCATACCCCGAACCAACCAAAX1X2CCAAAGGCAC-----)-3’
3’-d(-----GTGTGTGGCGACGATTGGGGTATGGGGCTTGGTTGGTTTGGGGTTTX3X4GTG-----)-5’
5’-d(-----CAACACAATAGTAGCCAAAACGCCCCCGAACCAACCAAAX5CCCAAAGGCAC-----)-3’
3’-d(-----GTTGTGTTATCATCGGTTTTGCGGGGGCTTGGTTGGTTTGGGGTTTX6CGTG-----)-5’
5’-d(-----CACACACCGCTGCTAACCCCATACCCCGAACCAACCAAAX7TATCCCCGCAC-----)-3’
a
Full sequence used is 70 nt long. The remaining sequence is depicted as ǦǦǦǦǦ.
To begin with, four donor-acceptor combinations were tested with tCO and
tCnitro incorporated both in the coding (X1,X2) and template strand (X3,X4) (Table
7 and Table 8). The FRET efficiencies obtained for these samples (17-21%; Table
8) confirm the integrity of the duplex since values are similar to those recorded in
a previous study for a 6 bp separation between tCO and tCnitro in B-DNA (using
fluorescence lifetimes: EFRET=20%)[40]. Next, TFAM was added to these
sequences, resulting in increases in FRET efficiencies between 8.3-15%. This
47
change in FRET indicates that TFAM causes significant structural changes around
the transcription start site.
We also determined the FRET efficiency of a sequence having a 7bp bubble
spanning the transcription start site (D4-ABUB, Table 8) and found a value of 65%.
This value is considerably larger than for the corresponding matching (D4-A4)
duplex with TFAM bound (EFRET,TFAM=29%), indicating that TFAM may induce
premelting around the transcription start site but that structural changes are not as
invasive as a 7bp unpaired region.
Finally, the structural effect of TFAM was investigated in a duplex with
mutations in the TFAM-binding sequence (DMUT-AMUT). No transcription could be
detected in vitro for sequences with mutations in this site. However, the same
change in FRET efficiency was observed for the mutated duplex as for the
corresponding wild-type sequence (D1-A1) upon addition of TFAM (Table 8).
Consequently, the structural changes induced by TFAM around the transcription
start site seem rather due to its capacity to bind and bend DNA in a sequenceindependent manner. Also a gel retardation experiment showed no difference in
TFAM binding to a wild type or mutated promoter sequence. These findings
indicate that TFAM not only induces structural alterations in the promoter
sequence but is probably also engaged in complex interactions with the rest of the
transcription machinery, for which the TFAM binding site is important.
Table 8 FRET efficiencies of tCO-tCnitro in four different DNA duplexes containing
the wild-type TFAM binding sequence as well as in a duplex with a mutated
TFAM binding site (DMUT-AMUT). Also a duplex with a 7 bp bubble spanning the
transcription start site was tested (D4-ABUB) to which no TFAM was added.
Duplex
tCO,a tCnitroa EFRET (%)b EFRET,TFAM (%)c
ΔEFRET (%)d
D1-A1
X1
X3
21 ± 0.2
31 ± 0.1
10 ± 0.2
D2-A2
X2
X4
17 ± 0.3
29 ± 0.4
12 ± 0.5
D3-A3
X4
X2
17 ± 0.3
32 ± 0.8
15 ± 0.9
D4-A4
X3
X1
21 ± 0.4
29 ± 0.4
8.3 ± 0.6
DMUT-AMUT
X5
X6
21 ± 0.2
30 ± 0.8
10 ± 0.8
D4-ABUB
X3
X7
65 ± 0.06
44 ± 0.4e
a
O
Position of tC or tCnitro in the corresponding sequences shown in Table 7. bFRET
efficiency upon addition of the corresponding buffer TFAM is dissolved in (except
for D4-ABUB). cSamples contain a ratio of 3:1 TFAM:DNA. dDifference in FRET
efficiency with and without TFAM present. eΔEFRET=EFRET,D4-ABUB-EFRET,D4-A4
Investigating the step-wise assembly of the transcription machinery
In paper VI, the role of the N-terminal extension (NTE) of Polrmt (mouse)
together with Tfam (mouse) was investigated in ensuring promoter-specific
transcription. A model is put forward in which Tfam recruits Polrmt to the
48
mitochondrial promoters, thereby relieving the inhibitory effect of the NTE of
Polrmt. In a next step, Tfb2m binds to the complex, allowing tight interactions
with the transcription start site and initiating transcription. This model is supported
by results from DNase I footprinting on the LSP promoter.
Also in this study we used the tCO-tCnitro FRET-pair to investigate whether
FRET-data would agree with the proposed model for transcription initiation. For
these experiments, the human HSP1, TFAM, POLRMT and TFB2M were applied,
also allowing the verification of the model at the mitochondrial promoter of
another organism. Duplex D4-A4 was used in these studies, with tCO-tCnitro
incorporated 6bp apart in opposite strands around the HSP1 transcription start site
as described above (Table 7 and Table 8). The change in FRET efficiency was
monitored upon addition of the three proteins in different orders (Table 9).
POLRMT alone has virtually no effect on the FRET efficiency (ΔEFRET=-0.20.6%). Subsequent addition of TFAM to this system only yields a small change in
FRET efficiency (ΔEFRET=2%) compared to the effect of TFAM without
POLMRT present (ΔEFRET=8.3%) (Table 9). Similarly, the presence of both
POLMRT and TFB2M leaves the FRET efficiency virtually unaltered
(ΔEFRET=0.4%). However, upon subsequent addition of TFAM to the latter case, a
change in FRET efficiency (12-13%) is observed which is larger than for TFAM
alone (8.3%). A possible explanation for these results could be that TFAM recruits
POLRMT to the promoter, with the latter suppressing non-specific interactions of
TFAM with the transcription start site. Next, binding of TFB2M allows melting
around the transcription start site and initiation of transcription. The structural
changes around the transcription start site suggested by our nucleobase FRET-pair
are in agreement with the model proposed above for the mouse LSP system.
Table 9 FRET efficiency of tCO-tCnitro in duplex D4-A4 (for sequences see Table 7
and Table 8 ) upon addition of POLRMT, TFAM and TFB2M (EFRET,protein) or
corresponding buffers (EFRET,buffer). Proteins were added in the order shown (1-3,
1-4 or 1 (only TFAM)).
DNA+protein
DNA:proteina
EFRET,buffer
EFRET,protein
ΔEFRET
c
(%)
(%)
(%)b
1) POLRMT
1:1.1
21 ± 0.2
21 ± 0.1
0.6 ± 0.2
2) TFAM
1:1.5
21 ± 0.2
23 ± 0.5
1.6 ± 0.5
3) TFAM
1:3
22 ± 0.1
24 ± 0.1
2.0 ± 0.1
1) POLRMT
1:1.1
21 ± 0.3
21 ± 0.2
-0.2 ± 0.4
2) TFB2M
1:1.1
21 ± 0.5
21 ± 0.5
0.4 ± 0.7
3) TFAM
1:1.5
21 ± 0.2
33 ± 0.5
12 ± 0.5
4) TFAM
1:3
21 ± 0.4
34 ± 0.2
13 ± 0.4
1) TFAM
1:3
21 ± 0.4
29 ± 0.4
8.3 ± 0.6
a
b
Relative molar ratios of DNA and proteins added. Difference in FRET efficiency
with and without proteins added.
49
50
5 Conclusion and outlook
The main focus of this thesis has been on fluorescent nucleobase analogues,
FBAs, which are probes for studying different aspects of DNA and RNA such as
their structure, dynamics and interactions with other molecules.
In the first part of this thesis, the characterization of photophysical and basemimicking properties of FBAs has been central. The spectroscopic
characterization of ten adenine analogues; AT, qA and the A7T-family; was
presented, of which the first two were also incorporated into DNA (paper I-III).
These adenine analogues show interesting variations in their emissive and basemimicking properties based on their structure. AT (paper I) shows a high
fluorescence quantum yield as a monomer (Φf=61% in water). Inside DNA, values
reach up to 21% for single strands and 5% for double strands. Its overall
brightness inside duplex DNA is roughly three times higher than the most
commonly used FBA, 2-AP, in a similar environment.[32,114] However, as for other
C8-modified purines, [116-118] AT moderately destabilizes the duplex structure.
Surprisingly, this modification also enables AT to base-pair with adenine besides
thymine.
In order to overcome the destabilizing effect caused by AT, a new series of 7modified triazole adenines, A7T-family (paper II), was developed, whose
photophysical properties as monomers have been discussed here. The members of
this new family were found to be significantly less emissive than AT in methanol.
Nevertheless, it will be interesting to investigate the fluorescent sister compound
of AT, 12a (Φf=5.2%) and the related 12b (Φf=4.4%) in DNA, since it is very
likely that they will be well accommodated in the duplex. A stabilizing effect has
namely been reported for other 7-substitited 8-aza-7-deazaadenines.[125] Moreover,
even though most FBAs are quenched inside nucleic acid systems, increasing
quantum yields have been reported for some analogues.[59] In the same line,
compound 20b may show promising fluorescence properties inside the base-stack
where it is more shielded from protic solvent molecules (Φf~21-24% in ACN or
DCM compared to 0.62% in methanol). However, hydrogen bonding to the
opposite base and to the triazole moiety protruding in the major groove may have
strong influences. Furthermore, it could be interesting to explore a four-ringed
variant of the A7T-family in which the carbon on position 5 of the triazole ring is
connected to the exocyclic amino moiety of adenine, thereby creating a more rigid
system, potentially with an increased emission.
The last adenine analogue studied in this work, qA (paper III) shows a
moderate quantum yield as a monomer (Φf=6.8% in water) and maximum values
of 5.8 % and 0.6 % in single- and double-stranded DNA, respectively. However,
51
inside duplex DNA, the overall brightness of qA is approximately ten times lower
than for AT. On the other hand, qA is a better isomorphic adenine analogue since it
shows specific base-pairing to thymine with the same preference for the type of
opposite base as natural adenine. Furthermore, qA on average stabilizes the DNA
duplexes, as has also been found for the ring-extended tricyclic cytosines tC[41]
and tCO[39]. The increase in melting temperature is dependent on the directly
neighboring bases and indicates that the two-ring extension is well accommodated
in the major groove. At the moment, new derivatives of qA are being
characterized which, according to calculations, are predicted to have increased
oscillator strengths. First results have shown improved photophysical properties,
in accordance with these predictions. This will hopefully lead to brighter
quadracyclic adenine analogues that retain the advantageous isomorphic properties
of qA.
In the interest of single-molecule types of studies, there is a large potential for
FBAs with an improved brightness and photostability. For that reason, it is
important to understand the photodegradation of these probes. In paper IV the
photostability of the already established tricyclic cytosine, tC, was therefore
characterized. It was found that its photoproduct, tC#, very likely is the sulfoxide
form of tC. Interestingly, tC# was found to significantly destabilize DNA
duplexes, causing partially denatured duplexes at room-temperature. This property
could have future applications in light-triggered therapeutics or in functional DNA
based nanodevices.[131-133]
In the second part of this work the tricyclic cytosine FRET-pair tCO-tCnitro[40]
was applied to study protein-DNA interactions in mitochondrial transcription. The
probes were incorporated around the transcription start site of the heavy strand
promoter (HSP1). Next, changes in FRET efficiency were monitored upon
addition of different components of the mitochondrial transcription machinery.
Since tCO and tCnitro are rigidly incorporated inside DNA, the observed signal
changes are expected to correspond to structural alterations of the duplex. In paper
V, the main focus was on the mammalian transcription factor A (TFAM), whereas
paper VI monitored the assembly of the whole transcription machinery. In both
studies, data obtained with the FRET-pair were in agreement with findings from
for example in vitro transcription and DNaseI footprinting. The FRET-pair is
being applied at the moment to study other protein-DNA interactions including
more detailed structural studies using the Matlab-based software FRETmatrix
developed by Preus et al.[108]. Also for these types of studies brighter and more
photostable FBAs will be a great asset. Furthermore, it will be highly
advantageous for the FBA FRET system, including FRETmatrix, if it can be
expanded with analogues of the other natural nucleobases so that incorporation
52
sites are not limited to cytosines. Potentially, the new qA analogues will provide
suitable adenine candidates for the task as FRET donor and/or acceptor.
Up till today, FBAs, including those discussed in this work, cannot compete
with external commercial dyes regarding optical properties such as brightness.
However, their main advantage is that they can be incorporated near or at the very
site of interest and therefore report on subtle structural changes in the nucleic acid
system, while causing minimal perturbations. This was nicely illustrated by our
DNA-protein interaction studies involving the FBA FRET-pair, tCO-tCnitro.
53
54
6 Acknowledgements
I would like to express my gratitude to the following people:
-First of all, to my supervisor, Marcus Wilhelmsson, for giving me the great
privilege of becoming your first PhD student. Thank you for the fantastic
guidance, support and discussions the past years. I consider myself very lucky.
-My co-supervisor Bo Albinsson, for interesting discussions, valuable advice and
for welcoming me into the Balb group.
-All my co-authors for interesting and exciting collaborations. Morten and Chris,
thank you for the nice discussions and advice.
-All current and former group members. Francois, Moa and Blaise: Thank you for
all the support and the nice atmosphere. Søren: thank you for all the help
throughout the years and the friendship. It has been a joy working with all of you.
-The Beuning group at Northeastern University for a fantastic spring in Boston,
especially Lisa and Philip for all the help.
-Former and current office mates for a very nice atmosphere.
-Jakob, Mélina and Tamas: thank you for friendship, great conversations and
support throughout the years!
-Current and former members of the climbing team for great Monday/Tuesday
evenings in and outside of fabben: Johanna, Peter, Jong-Ah, Maria A., Johan,
Louise and Fabian.
-Everyone at the division of Physical chemistry, past and present, for making
Sweden my second home.
-Blaise, Moa, Chris, Jakob and Basti for reading this thesis.
-Friends, in Sweden, Germany and back home in Belgium. Julie, thank you for our
fantastic friendship.
-My family for unconditional love and support.
-Basti, for loving, caring and (a lot of) listening.
55
56
7 References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
Weaver, R.F. (2005) Molecular biology 3ed. McGraw-Hill, New York,
NY.
Dahm, R. (2005) Friedrich Miescher and the discovery of DNA. Dev.
Biol., 278, 274-288.
Watson, J.D. and Crick, F.H.C. (1953) Molecular structure of nucleic
acids - A structure for deoxyribose nulceic acid. Nature, 171, 737-738.
Reese, C.B. (2005) Oligo- and poly-nucleotides: 50 years of chemical
synthesis. Org. Biomol. Chem., 3, 3851-3868.
atdbio: http://www.atdbio.com/.
Lander, E.S., Int Human Genome Sequencing, C., Linton, L.M., Birren,
B., Nusbaum, C., Zody, M.C., Baldwin, J., Devon, K., Dewar, K., Doyle,
M. et al. (2001) Initial sequencing and analysis of the human genome.
Nature, 409, 860-921.
Mattick, J.S. and Makunin, I.V. (2006) Non-coding RNA. Human
Molecular Genetics, 15, R17-R29.
Bartel, D.P. (2009) MicroRNAs: Target Recognition and Regulatory
Functions. Cell, 136, 215-233.
Montange, R.K. and Batey, R.T. (2008) Riboswitches: Emerging themes
in RNA structure and function. Annu. Rev. Biophys, 37, 117-133.
Serganov, A. and Patel, D.J. (2007) Ribozymes, riboswitches and
beyond: regulation of gene expression without proteins. Nat Rev Genet,
8, 776-790.
Seeman, N.C. (2010) Structural DNA Nanotechnology: Growing Along
with Nano Letters. Nano Lett., 10, 1971-1978.
Modi, S., Bhatia, D., Simmel, F.C. and Krishnan, Y. (2010) Structural
DNA Nanotechnology: From Bases to Bricks, From Structure to
Function. J. Phys. Chem. Lett., 1, 1994-2005.
Pinheiro, A.V., Han, D.R., Shih, W.M. and Yan, H. (2011) Challenges
and opportunities for structural DNA nanotechnology. Nat. Nanotechnol.,
6, 763-772.
Holbrook, S.R. (2008) Structural principles from large RNAs. Annu. Rev.
Biophys, 37, 445-464.
Foster, M.P., McElroy, C.A. and Amero, C.D. (2007) Solution NMR of
large molecules and assemblies. Biochemistry, 46, 331-340.
Lakowicz, J.R. (2006) Principles of Fluorescence Spectroscopy. 3 ed.
Springer, New York, NY.
Onidas, D., Markovitsi, D., Marguet, S., Sharonov, A. and Gustavsson,
T. (2002) Fluorescence properties of DNA nucleosides and nucleotides:
A refined steady-state and femtosecond investigation. J. Phys. Chem. B,
106, 11367-11374.
Callis, P.R. (1983) Electronic states and luminescence of nucleic-acid
systems. Annu. Rev. Phys. Chem., 34, 329-357.
57
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
Serrano-Andres, L. and Merchan, M. (2009) Are the five natural
DNA/RNA base monomers a good choice from natural selection? A
photochemical perspective. J. Photochem. Photobiol., C, 10, 21-32.
Waring, M.J. (1965) Complex formation between ethidium bromide and
nucleic acids. J. Mol. Biol., 13, 269-282.
Larsen, T.A., Goodsell, D.S., Cascio, D., Grzeskowiak, K. and
Dickerson, R.E. (1989) The structure of DAPI bound to DNA. J. Biomol.
Struct. Dyn., 7, 477-491.
Fornander, L.H., Wu, L.S., Billeter, M., Lincoln, P. and Norden, B.
(2013) Minor-Groove Binding Drugs: Where Is the Second Hoechst
33258 Molecule? J. Phys. Chem. B, 117, 5820-5830.
Larsson, A., Carlsson, C., Jonsson, M. and Albinsson, B. (1994)
Characterization of the binding of the fluorescent dyes YO and YOYO to
DNA by polarized-light spectroscopy. J. Am. Chem. Soc., 116, 84598465.
Ha, T. (2001) Single-molecule fluorescence methods for the study of
nucleic acids. Curr. Opin. Struct. Biol., 11, 287-292.
Iqbal, A., Arslan, S., Okumus, B., Wilson, T.J., Giraud, G., Norman,
D.G., Ha, T. and Lilley, D.M.J. (2008) Orientation dependence in
fluorescent energy transfer between Cy3 and Cy5 terminally attached to
double-stranded nuclelic acids. Proc. Natl. Acad. Sci. U. S. A., 105,
11176-11181.
Mollova, E.T. (2002) Single-molecule fluorescence of nucleic acids.
Curr. Opin. Chem. Biol., 6, 823-828.
Hawkins, M.E. (2001) Fluorescent pteridine nucleoside analogs - A
window on DNA interactions. Cell Biochem. Biophys., 34, 257-281.
Rist, M.J. and Marino, J.P. (2002) Fluorescent nucleotide base analogs as
probes of nucleic acid structure, dynamics and interactions. Curr. Org.
Chem., 6, 775-793.
Sinkeldam, R.W., Greco, N.J. and Tor, Y. (2010) Fluorescent Analogs of
Biomolecular Building Blocks: Design, Properties, and Applications.
Chem. Rev., 110, 2579-2619.
Wilhelmsson, L.M. (2010) Fluorescent nucleic acid base analogues. Q.
Rev. Biophys., 43, 159-183.
Wilson, J.N. and Kool, E.T. (2006) Fluorescent DNA base replacements:
reporters and sensors for biological systems. Org. Biomol. Chem., 4,
4265-4274.
Ward, D.C., Reich, E, Stryer, L. (1969) Fluorescence Studies of
Nucleotides and Polynucleotides. J. Biol. Chem., 244, 1228-1237.
Xu, D.G., Evans, K.O. and Nordlund, T.M. (1994) Melting and
premelting transitions of an oligomer measured by DNA-base
fluorescence and absorption. Biochemistry, 33, 9592-9599.
Sowers, L.C., Boulard, Y. and Fazakerley, G.V. (2000) Multiple
structures for the 2-aminopurine-cytosine mispair. Biochemistry, 39,
7613-7620.
Sowers, L.C., Fazakerley, G.V., Eritja, R., Kaplan, B.E. and Goodman,
M.F. (1986) Base-pairing and mutagenesis - observation of a protonated
58
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
base pair between 2-aminopurine and cytosine in an oligonucleotide by
proton NMR. Proc. Natl. Acad. Sci. U. S. A., 83, 5434-5438.
Law, S.M., Eritja, R., Goodman, M.F. and Breslauer, K.J. (1996)
Spectroscopic and calorimetric characterizations of DNA duplexes
containing 2-aminopurine. Biochemistry, 35, 12329-12337.
Dyrager, C., Börjesson, K., Diner, P., Elf, A., Albinsson, B.,
Wilhelmsson, L.M. and Grøtli, M. (2009) Synthesis and Photophysical
Characterisation of Fluorescent 8-(1H-1,2,3-Triazol-4-yl)adenosine
Derivatives. Eur. J. Org. Chem., 1515-1521.
Sandin, P., Wilhelmsson, L.M., Lincoln, P., Powers, V.E.C., Brown, T.
and Albinsson, B. (2005) Fluorescent properties of DNA base analogue
tC upon incorporation into DNA - negligible influence of neighbouring
bases on fluorescence quantum yield. Nucleic Acids Res., 33, 5019-5025.
Sandin, P., Börjesson, K., Li, H., Mårtensson, J., Brown, T.,
Wilhelmsson, L.M. and Albinsson, B. (2008) Characterization and use of
an unprecedentedly bright and structurally non-perturbing fluorescent
DNA base analogue. Nucleic Acids Res., 36, 157-167.
Börjesson, K., Preus, S., El-Sagheer, A.H., Brown, T., Albinsson, B. and
Wilhelmsson, L.M. (2009) Nucleic Acid Base Analog FRET-Pair
Facilitating Detailed Structural Measurements in Nucleic Acid
Containing Systems. J. Am. Chem. Soc., 131, 4288-4293.
Engman, K.C., Sandin, P., Osborne, S., Brown, T., Billeter, M., Lincoln,
P., Nordén, B., Albinsson, B. and Wilhelmsson, L.M. (2004) DNA
adopts normal B-form upon incorporation of highly fluorescent DNA
base analogue tC: NMR structure and UV-Vis spectroscopy
characterization. Nucleic Acids Res., 32, 5087-5095.
Preus, S., Börjesson, K., Kilså, K., Albinsson, B. and Wilhelmsson, L.M.
(2010) Characterization of Nucleobase Analogue FRET Acceptor tCnitro.
J. Phys. Chem. B, 114, 1050-1056.
Avery, O.T., MacLeod, C.M. and McCarty, M. (1944) Studies on the
chemical nature of the substance inducing transformation of
pneumococcal types induction of transformation by a desoxyribonucleic
acid fraction isolated from Pneumococcus type III. J. Exp. Med., 79, 137158.
Chargaff, E. (1950) Chemical specificity of nucleic acids and mechanism
of their enzymatic degradation. Experientia, 6, 201-209.
Franklin, R.E. and Gosling, R.G. (1953) Molecular configuration in
sodium thymonucleate. Nature, 171, 740-741.
Wilkins, M.H.F., Stokes, A.R. and Wilson, H.R. (1953) Molecular
structure of deoxypentose nucleic acids. Nature, 171, 738-740.
Garrett, R.H. and Grisham, C.M. (2005) Biochemistry. 3 ed. Thomson
Brooks/Cole, Belmont, CA.
Bloomfield, V.A., Crothers, D.M., Tinoco, I.Jr. (2000) Nucleic acids:
Structures, properties and functions. University Science books, Sausalito,
CA.
Stofer, E. and Lavery, R. (1994) Measuring the Geometry of DNA
Grooves. Biopolymers, 34, 337-346.
59
[50]
[51]
[52]
[53]
[54]
[55]
[56]
[57]
[58]
[59]
[60]
[61]
[62]
[63]
Choi, J. and Majima, T. (2011) Conformational changes of non-B DNA.
Chem. Soc. Rev., 40, 5893-5909.
Woller, J.G., Hannestad, J.K. and Albinsson, B. (2013) Self-Assembled
Nanoscale DNA-Porphyrin Complex for Artificial Light Harvesting. J.
Am. Chem. Soc., 135, 2759-2768.
Unruh, J.R., Gokulrangan, G., Lushington, G.H., Johnson, C.K. and
Wilson, G.S. (2005) Orientational dynamics and dye-DNA interactions in
a dye-labeled DNA aptamer. Biophys. J., 88, 3455-3465.
Norman, D.G., Grainger, R.J., Uhrin, D. and Lilley, D.M.J. (2000)
Location of cyanine-3 on double-stranded DNA: Importance for
fluorescence resonance energy transfer studies. Biochemistry, 39, 63176324.
Singh, M.P., Jennings, T.L. and Strouse, G.F. (2009) Tracking Spatial
Disorder in an Optical Ruler by Time-Resolved NSET. J. Phys. Chem. B,
113, 552-558.
Ouellet, J., Schorr, S., Iqbal, A., Wilson, T.J. and Lilley, D.M.J. (2011)
Orientation of Cyanine Fluorophores Terminally Attached to DNA via
Long, Flexible Tethers. Biophys. J., 101, 1148-1154.
Iqbal, A., Wang, L., Thompson, K.C., Lilley, D.M.J. and Norman, D.G.
(2008) The structure of cyanine 5 terminally attached to double-stranded
DNA: Implications for FRET studies. Biochemistry, 47, 7857-7862.
Neubauer, H., Gaiko, N., Berger, S., Schaffer, J., Eggeling, C., Tuma, J.,
Verdier, L., Seidel, C.A.M., Griesinger, C. and Volkmer, A. (2007)
Orientational and dynamical heterogeneity of rhodamine 6G terminally
attached to a DNA helix revealed by NMR and single-molecule
fluorescence spectroscopy. J. Am. Chem. Soc., 129, 12746-12755.
Okamoto, A., Saito, Y. and Saito, I. (2005) Design of base-discriminating
fluorescent nucleosides. J. Photochem. Photobiol., C, 6, 108-122.
Zhao, Y., Knee, J.L. and Baranger, A.M. (2008) Characterization of two
adenosine analogs as fluorescence probes in RNA. Bioorg. Chem., 36,
271-277.
Driscoll, S.L., Hawkins, M.E., Balis, F.M., Pfleiderer, W. and Laws,
W.R. (1997) Fluorescence properties of a new guanosine analog
incorporated into small oligonucleotides. Biophys. J., 73, 3277-3286.
Hawkins, M.E., Pfleiderer, W., Jungmann, O. and Balis, F.M. (2001)
Synthesis and fluorescence characterization of pteridine adenosine
nucleoside analogs for DNA incorporation. Anal. Biochem., 298, 231240.
Hawkins, M.E., Pfleiderer, W., Balis, F.M., Porter, D. and Knutson, J.R.
(1997) Fluorescence properties of pteridine nucleoside analogs as
monomers and incorporated into oligonucleotides. Anal. Biochem., 244,
86-95.
Greco, N.J. and Tor, Y. (2005) Simple fluorescent pyrimidine analogues
detect the presence of DNA abasic sites. J. Am. Chem. Soc., 127, 1078410785.
60
[64]
[65]
[66]
[67]
[68]
[69]
[70]
[71]
[72]
[73]
[74]
[75]
[76]
[77]
Srivatsan, S.G., Tor Y. (2007) Fluorescent Pyrimidine Ribonucleotide:
Synthesis, Enzymatic Incorporation, and Utilization. J. Am. Chem. Soc.,
129, 2044-2053.
Srivatsan, S.G., Greco, N.J. and Tor, Y. (2008) A highly emissive
fluorescent nucleoside that signals the activity of toxic ribosomeinactivating proteins. Angew. Chem., Int. Ed. Engl., 47, 6661-6665.
Srivatsan, S.G., Weizman, H. and Tor, Y. (2008) A highly fluorescent
nucleoside analog based on thieno 3,4-d pyrimidine senses mismatched
pairing. Org. Biomol. Chem., 6, 1334-1338.
Sinkeldam, R.W., Hopkins, P.A. and Tor, Y. (2012) Modified 6-Aza
Uridines: Highly Emissive pH-Sensitive Fluorescent Nucleosides.
Chemphyschem, 13, 3350-3356.
Shin, D., Sinkeldam, R.W. and Tor, Y. (2011) Emissive RNA alphabet.
J. Am. Chem. Soc., 133, 14912-14915.
Ben Gaied, N., Glasser, N., Ramalanjaona, N., Beltz, H., Wolff, P.,
Marquet, R., Burger, A. and Mely, Y. (2005) 8-vinyl-deoxyadenosine, an
alternative fluorescent nucleoside analog to 2'-deoxyribosyl-2aminopurine with improved properties. Nucleic Acids Res., 33, 10311039.
Krueger, A.T. and Kool, E.T. (2008) Fluorescence of size-expanded
DNA bases: Reporting on DNA sequence and structure with an unnatural
genetic set. J. Am. Chem. Soc., 130, 3989-3999.
Liu, H.B., Gao, J.M. and Kool, E.T. (2005) Helix-forming properties of
size-expanded DNA, an alternative four-base genetic form. J. Am. Chem.
Soc., 127, 1396-1402.
Gao, J.M., Liu, H.B. and Kool, E.T. (2004) Expanded-size bases in
naturally sized DNA: Evaluation of steric effects in Watson-Crick
pairing. J. Am. Chem. Soc., 126, 11826-11831.
Berry, D.A., Jung, K.Y., Wise, D.S., Sercel, A.D., Pearson, W.H.,
Mackie, H., Randolph, J.B. and Somers, R.L. (2004) Pyrrolo-dC and
pyrrolo-C: fluorescent analogs of cytidine and 2 '-deoxycytidine for the
study of oligonucleotides. Tetrahedron Lett., 45, 2457-2461.
Noé, M.S., Ríos, A.C. and Tor, Y. (2012) Design, Synthesis, and
Spectroscopic Properties of Extended and Fused Pyrrolo-dC and PyrroloC Analogs. Org. Lett., 14, 3150-3153.
Seela, F. and Sirivolu, V.R. (2008) Pyrrolo-dC oligonucleotides bearing
alkynyl side chains with terminal triple bonds: synthesis, base pairing and
fluorescent dye conjugates prepared by the azide-alkyne "click" reaction.
Org. Biomol. Chem., 6, 1674-1687.
Suchy, M. and Hudson, R.H.E. (2014) Pyrimidine-Fused Heterocyclic
Frameworks Based on an N4-Arylcytosine Scaffold: Synthesis,
Characterization, and PNA Oligomerization of the Fluorescent Cytosine
Analogue 5,6-BenzopC. J. Org. Chem., 79, 3336-3347.
Wojciechowski, F. and Hudson, R.H.E. (2008) Fluorescence and
hybridization properties of peptide nucleic acid containing a substituted
phenylpyrrolocytosine designed to engage guanine with an additional Hbond. J. Am. Chem. Soc., 130, 12574-12575.
61
[78]
[79]
[80]
[81]
[82]
[83]
[84]
[85]
[86]
[87]
[88]
[89]
[90]
[91]
Rachofsky, E.L., Osman, R. and Ross, J.B.A. (2001) Probing structure
and dynamics of DNA with 2-aminopurine: Effects of local environment
on fluorescence. Biochemistry, 40, 946-956.
Augustyn, K.E., Wojtuszewski, K., Hawkins, M.E., Knutson, J.R. and
Mukerji, I. (2006) Examination of the premelting transition of DNA Atracts using a fluorescent adenosine analogue. Biochemistry, 45, 50395047.
Stivers, J.T. (1998) 2-aminopurine fluorescence studies of base stacking
interactions at abasic sites in DNA: metal-ion and base sequence effects.
Nucleic Acids Res., 26, 3837-3844.
Guest, C.R., Hochstrasser, R.A., Sowers, L.C. and Millar, D.P. (1991)
Dynamics of mismatched base-pairs in DNA. Biochemistry, 30, 32713279.
Bandwar, R.P. and Patel, S.S. (2001) Peculiar 2-aminopurine
fluorescence monitors the dynamics of open complex formation by
bacteriophage T7 RNA polymerase. J. Biol. Chem., 276, 14075-14082.
Deprez, E., Tauc, P., Leh, H., Mouscadet, J.F., Auclair, C., Hawkins,
M.E. and Brochon, J.C. (2001) DNA binding induces dissociation of the
multimeric form of HIV-1 integrase: A time-resolved fluorescence
anisotropy study. Proc. Natl. Acad. Sci. U. S. A., 98, 10090-10095.
Hawkins, M.E., Pfleiderer, W., Mazumder, A., Pommier, Y.G. and Falls,
F.M. (1995) Incorporation of a fluorescent guanosine analog into
oligonucleotides and its application to a real-time assay for the HIV-1
Integrase 3'-processing reaction. Nucleic Acids Res., 23, 2872-2880.
Sandin, P., Tumpane, J., Börjesson, K., Wilhelmsson, L.M., Brown, T.,
Nordén, B., Albinsson, B. and Lincoln, P. (2009) Thermodynamic
Aspects of DNA Nanoconstruct Stability and Design. J. Phys. Chem. C,
113, 5941-5946.
Lemay, J.F., Penedo, J.C., Tremblay, R., Lilley, D.M.J. and Lafontaine,
D.A. (2006) Folding of the adenine riboswitch. Chem. Biol., 13, 857-868.
Hennelly, S.P. and Sanbonmatsu, K.Y. (2011) Tertiary contacts control
switching of the SAM-I riboswitch. Nucleic Acids Res., 39, 2416-2431.
Wahba, A.S., Azizi, F., Deleavey, G.F., Brown, C., Robert, F., Carrier,
M., Kalota, A., Gewirtz, A.M., Pelletier, J., Hudson, R.H.E. et al. (2011)
Phenylpyrrolocytosine as an Unobtrusive Base Modification for
Monitoring Activity and Cellular Trafficking of siRNA. ACS Chemical
Biology, 6, 912-919.
Yi-Brunozzi, H.Y., Stephens, O.M. and Beal, P.A. (2001)
Conformational changes that occur during an RNA-editing adenosine
deamination reaction. J. Biol. Chem., 276, 37827-37833.
Stephens, O.M., Yi-Brunozzi, H.Y. and Beal, P.A. (2000) Analysis of the
RNA-editing reaction of ADAR2 with structural and fluorescent
analogues of the GluR-B R/G editing site. Biochemistry, 39, 1224312251.
Gilbert, A. and Baggott, J.E. (1991) Essentials of molecular
photochemistry. Blackwell Scientific Publications, Oxford.
62
[92]
[93]
[94]
[95]
[96]
[97]
[98]
[99]
[100]
[101]
[102]
[103]
[104]
[105]
[106]
[107]
[108]
[109]
[110]
Atkins, P.P.W. and De Paula, J. (2002) Atkins' Physical Chemistry.
Oxford University Press, Incorporated, New York.
Parson, W.W. (2007) Modern Optical Spectroscopy. Springer-Verlag,
Heidelberg.
Nordén, B., Rodger, A., Dafforn, T. (2010) Linear Dichrosim and
Circular Dichrosim: A textbook on polarized-light spectroscopy. The
Royal Society of Chemistry, Cambridge, UK.
Valeur, B. (2001) Molecular Fluorescence: Principles and Applications.
Wiley-VCH, Weinheim.
Strickler, S.J. and Berg, R.A. (1962) Relationship between absorption
intensity and fluorescence lifetime of molecules. J. Chem. Phys., 37, 814822.
Fiserova, E. and Kubala, M. (2012) Mean fluorescence lifetime and its
error. J. Lumin., 132, 2059-2064.
Förster, T. (1948) Zwischenmolekulare energiewanderung und
fluoreszenz. Ann. Phys.-Berlin, 2, 55-75.
Förster,
T.
(1946)
Energiewanderung
und
Fluoreszenz.
Naturwissenschaften, 33, 166-175.
Dale, R.E., Eisinger, J. and Blumberg, W.E. (1979) Orientational
freedom of molecular probes - orientation factor in intra-molecular
energy-transfer. Biophys. J., 26, 161-193.
Stryer, L. (1978) Fluorescence energy-transfer as a spectroscopic ruler.
Annu. Rev. Biochem., 47, 819-846.
Stryer, L. and Haugland, R.P. (1967) Energy transfer - a spectroscopic
ruler. Proc. Natl. Acad. Sci. U. S. A., 58, 719-726.
Preus, S. and Wilhelmsson, L.M. (2012) Advances in Quantitative
FRET-Based Methods for Studying Nucleic Acids. ChemBioChem, 13,
1990-2001.
Wozniak, A.K., Schroder, G.F., Grubmuller, H., Seidel, C.A.M. and
Oesterhelt, F. (2008) Single-molecule FRET measures bends and kinks in
DNA. Proc. Natl. Acad. Sci. U. S. A., 105, 18337-18342.
Roy, R., Hohng, S. and Ha, T. (2008) A practical guide to singlemolecule FRET. Nat. Methods, 5, 507-516.
Lewis, F.D., Zhang, L.G. and Zuo, X.B. (2005) Orientation control of
fluorescence resonance energy transfer using DNA as a helical scaffold.
J. Am. Chem. Soc., 127, 10002-10003.
Kato, T., Kashida, H., Kishida, H., Yada, H., Okamoto, H. and Asanuma,
H. (2013) Development of a Robust Model System of FRET using Base
Surrogates Tethering Fluorophores for Strict Control of Their Position
and Orientation within DNA Duplex. J. Am. Chem. Soc., 135, 741-750.
Preus, S., Kilsa, K., Miannay, F.A., Albinsson, B. and Wilhelmsson,
L.M. (2013) FRETmatrix: a general methodology for the simulation and
analysis of FRET in nucleic acids. Nucleic Acids Res., 41.
Tinoco, I. (1960) Hypochromism in polynucleotides. J. Am. Chem. Soc.,
82, 4785-4790.
Rhodes, W. (1961) Hypochromism and other spectral properties of
helical polynucleotides. J. Am. Chem. Soc., 83, 3609-3617.
63
[111]
[112]
[113]
[114]
[115]
[116]
[117]
[118]
[119]
[120]
[121]
[122]
[123]
[124]
[125]
Chipem, F.A.S., Mishra, A. and Krishnamoorthy, G. (2012) The role of
hydrogen bonding in excited state intramolecular charge transfer.
Physical Chemistry Chemical Physics, 14, 8775-8790.
Lakowicz, J.R. (2006), Principles of Fluorescence Spectroscopy. 3 ed.
Springer, New York, pp. 205-235.
Albinsson, B. (1997) Dual fluorescence from N6,N6-dimethyladenosine.
J. Am. Chem. Soc., 119, 6369-6375.
Smagowicz, J. and Wierzchowski, K.L. (1974) Lowest excited states of
2-aminopurine. J. Lumin., 8, 210-232.
Nadler, A., Strohmeier, J. and Diederichsen, U. (2011) 8-Vinyl-2 'deoxyguanosine as a Fluorescent 2 '-Deoxyguanosine Mimic for
Investigating DNA Hybridization and Topology. Angew. Chem.-Int.
Edit., 50, 5392-5396.
Catalanotti, B., Galeone, A., Gomez-Paloma, L., Mayol, L. and Pepe, A.
(2000) 2 '-deoxy-8-(propyn-1-yl)adenosine-containing oligonucleotides:
Effects on stability of duplex and quadruples structures. Bioorg. Med.
Chem. Lett., 10, 2005-2009.
Rankin, K.M., Sproviero, M., Rankin, K., Sharma, P., Wetmore, S.D. and
Manderville, R.A. (2012) C 8-Heteroaryl-2′-deoxyguanosine Adducts as
Conformational Fluorescent Probes in the Nar I Recognition Sequence.
The Journal of organic chemistry, 77, 10498-10508.
Eason, R.G., Burkhardt, D.M., Phillips, S.J., Smith, D.P. and David, S.S.
(1996) Synthesis and characterization of 8-methoxy-2'-deoxyadenosinecontaining oligonucleotides to probe the syn glycosidic conformation of
2'-deoxyadenosine within DNA. Nucleic Acids Res., 24, 890-897.
Millen, A.L., McLaughlin, C.K., Sun, K.M., Manderville, R.A. and
Wetmore, S.D. (2008) Computational and experimental evidence for the
structural preference of phenolic C-8 purine adducts. J. Phys. Chem. A,
112, 3742-3753.
Tavale, S.S. and Sobell, H.M. (1970) Crystal and molecular structure of
8-bromoguanosine and 8-bromoadenosine, 2 purine nucleosides in syn
conformation. J. Mol. Biol., 48, 109-&.
Millen, A.L., Manderville, R.A. and Wetmore, S.D. (2010)
Conformational Flexibility of C8-Phenoxyl-2 '-deoxyguanosine
Nucleotide Adducts. J. Phys. Chem. B, 114, 4373-4382.
Barkley, M.D. and Zimm, B.H. (1979) Theory of twisting and bending of
chain macromolecules - analysis of the fluorescence depolarization of
DNA. J. Chem. Phys., 70, 2991-3007.
Seela, F. and Thomas, H. (1995) Duplex stabilization of DNA oligonucleotides containing 7-substituted 7-deazaadenines. Helv. Chim.
Acta, 78, 94-108.
Seela, F. and Zulauf, M. (1998) 7-deazaadenine-DNA: Bulky 7-iodo
substituents or hydrophobic 7-hexynyl chains are well accommodated in
the major groove of oligonucleotide duplexes. Chem.-Eur. J., 4, 17811790.
Seela, F. and Zulauf, M. (1999) Synthesis of oligonucleotides containing
pyrazolo 3,4-d pyrimidines: The influence of 7-substituted 8-aza-7-
64
[126]
[127]
[128]
[129]
[130]
[131]
[132]
[133]
[134]
[135]
[136]
[137]
[138]
[139]
deazaadenines on the duplex structure and stability. J. Chem. Soc., Perkin
Trans. 1, 479-488.
Somsen, O.J.G., Hoek, v.A. and Amerongen, v.H. (2005) Fluorescence
quenching of 2-aminopurine in dinucleotides. Chem. Phys. Lett., 402, 6165.
Shafirovich, V., Dourandin, A., Huang, W.D., Luneva, N.P. and
Geacintov, N.E. (2000) Electron transfer at a distance induced by siteselective photoionization of 2-aminopurine in oligonucleotides and
investigated by transient absorption techniques. Phys. Chem. Chem.
Phys., 2, 4399-4408.
Kenfack, C.A., Burger, A. and Mely, Y. (2006) Excited-state properties
and transitions of fluorescent 8-vinyl adenosine in DNA. J. Phys. Chem.
B, 110, 26327-26336.
Zhang, X. and Wadkins, R.M. (2009) DNA Hairpins Containing the
Cytidine Analog Pyrrolo-dC: Structural, Thermodynamic, and
Spectroscopic Studies. Biophys. J., 96, 1884-1891.
Singleton, S.F., Roca, A.I., Lee, A.M. and Xiao, J. (2007) Probing the
structure of RecA-DNA filaments. Advantages of a fluorescent guanine
analog. Tetrahedron, 63, 3553-3566.
Brieke, C., Rohrbach, F., Gottschalk, A., Mayer, G. and Heckel, A.
(2012) Light-Controlled Tools. Angew. Chem.-Int. Edit., 51, 8446-8476.
Wang, G.J. and Zhang, J. (2012) Photoresponsive molecular switches for
biotechnology. J. Photochem. Photobiol., C, 13, 299-309.
Ceo, L.M. and Koh, J.T. (2012) Photocaged DNA Provides New Levels
of Transcription Control. ChemBioChem, 13, 511-513.
Falkenberg, M., Larsson, N.G. and Gustafsson, C.M. (2007), Annu. Rev.
Biochem., Vol. 76, pp. 679-699.
Montoya, J., Christianson, T., Levens, D., Rabinowitz, M. and Attardi, G.
(1982) Identification of initiation sites for heavy-strand and light-strand
transcription in human mitochondrial-DNA. Proc. Natl. Acad. Sci. U. S.
A., 79, 7195-7199.
Litonin, D., Sologub, M., Shi, Y.H., Savkina, M., Anikin, M.,
Falkenberg, M., Gustafsson, C.M. and Temiakov, D. (2010) Human
mitochondrial transcription revisited: only TFAM and TFB2M are
required for transcription of the mitochondrial genes in vitro. J. Biol.
Chem., 285, 18129-18133.
Falkenberg, M., Gaspari, M., Rantanen, A., Trifunovic, A., Larsson, N.G.
and Gustafsson, C.M. (2002) Mitochondrial transcription factors B1 and
B2 activate transcription of human mtDNA. Nat. Genet., 31, 289-294.
Kanki, T., Ohgaki, K., Gaspari, M., Gustafsson, C.M., Fukuoh, A.,
Sasaki, N., Hamasaki, N. and Kang, D.C. (2004) Architectural role of
mitochondrial transcription factor A in maintenance of human
mitochondrial DNA. Mol. Cell. Biol., 24, 9823-9834.
Shutt, T.E., Lodeiro, M.F., Cotney, J., Cameron, C.E. and Shadel, G.S.
(2010) Core human mitochondrial transcription apparatus is a regulated
two-component system in vitro. Proc. Natl. Acad. Sci. U. S. A., 107,
12133-12138.
65
[140]
[141]
[142]
[143]
[144]
[145]
Fisher, R.P., Topper, J.N. and Clayton, D.A. (1987) Promoter selection in
human mitochondria involves binding of a transcription factor to
orientation-independent upstream regulatory elements. Cell, 50, 247-258.
Gaspari, M., Falkenberg, M., Larsson, N.G. and Gustafsson, C.M. (2004)
The mitochondrial RNA polymerase contributes critically to promoter
specificity in mammalian cells. EMBO J., 23, 4606-4614.
Fisher, R.P., Lisowsky, T., Parisi, M.A. and Clayton, D.A. (1992) DNA
wrapping and bending by a mitochondrial high mobility group-like
transcriptional activator protein. J. Biol. Chem., 267, 3358-3367.
Diffley, J.F.X. and Stillman, B. (1991) A close relative of the nuclear,
chromosomal high-mobility group protein HMG1 in yeast mitochondria.
Proc. Natl. Acad. Sci. U. S. A., 88, 7864-7868.
Parisi, M.A., Xu, B.J. and Clayton, D.A. (1993) A human mitochondrial
transcriptional activator can functionally replace a yeast mitochondrial
HMG-box protein both in vivo and in vitro. Mol. Cell. Biol., 13, 19511961.
Kaufman, B.A., Durisic, N., Mativetsky, J.M., Costantino, S., Hancock,
M.A., Grutter, P. and Shoubridge, E.A. (2007) The mitochondrial
transcription factor TFAM coordinates the assembly of multiple DNA
molecules into nucleoid-like structures. Mol. Biol. Cell, 18, 3225-3236.
66